Depression is a choice

I enjoyed Matt Yglesias’ suggestion that depressions are merely a technical problem that will go away once the obsolescence of cash eliminates the zero lower bound on interest rates, and Ryan Avent’s rejoinder. Although I’ve toyed with Yglesias’ view myself, I think that Avent has the better of the argument when he characterizes our current policy impotence as reflecting behavioral rather than technical constraints. We don’t lack for technical means to counter people’s self-defeating impulse to hoard cash and safe financial assets. On the contrary, we have a whole cornucopia of options! The squabbling that has preoccupied me lately, between market monetarists and post-Keynesians and mainstream saltwater economists, is an argument over which of many not-necessarily-mutually-exclusive options would most perfectly address address this not-really-challenging problem.

We are in a depression, but not because we don’t know how to remedy the problem. We are in a depression because it is our revealed preference, as a polity, not to remedy the problem. We are choosing continued depression because we prefer it to the alternatives.

Usually, economists are admirably catholic about the preferences of the objects they study. They infer desire by observing behavior, listening to what people do more than to what they say. But with respect to national polities, macroeconomists presume the existence of an overwhelming preference for GDP growth and full employment that simply does not exist. They act as though any other set of preferences would be unreasonable, unthinkable.

But the preferences of developed, aging polities — first Japan, now the United States and Europe — are obvious to a dispassionate observer. Their overwhelming priority is to protect the purchasing power of incumbent creditors. That’s it. That’s everything. All other considerations are secondary. These preferences are reflected in what the polities do, how they behave. They swoop in with incredible speed and force to bail out the financial sectors in which creditors are invested, trampling over prior norms and laws as necessary. The same preferences are reflected in what the polities omit to do. They do not pursue monetary policy with sufficient force to ensure expenditure growth even at risk of inflation. They do not purse fiscal policy with sufficient force to ensure employment even at risk of inflation. They remain forever vigilant that neither monetary ease nor fiscal profligacy engender inflation. The tepid policy experiments that are occasionally embarked upon they sabotage at the very first hint of inflation. The purchasing power of holders of nominal debt must not be put at risk. That is the overriding preference, in context of which observed behavior is rational.

I am often told that this is absurd because, after all, wouldn’t creditors be better off in a booming economy than in a depressed one? In a depression, creditors may not face unexpected inflation, sure. But they also earn next to nothing on their money, sometimes even a bit less than nothing in real terms. “Financial repression! Savers are being squeezed!” In a boom, they would enjoy positive interest rates.

That’s true. But the revealed preference of the polity is not balanced. It is not some cartoonish capitalist-class conspiracy story, where the goal is to maximize the wealth of exploiters. The revealed preference of the polity is to resist losses for incumbent creditors much more than it is to seek gains. In a world of perfect certainty, given a choice between recession and boom, the polity would choose boom. But in the real world, the polity faces great uncertainty. The policies that might engender a boom are not guaranteed to succeed. They carry with them a short-to-medium-term risk of inflation, perhaps even a significant inflation if things don’t go as planned. The polity prefers inaction to bearing this risk.

This preference is not at all difficult to understand. The ailing developed economies are plutocratic democracies. “The people” do have power, but influence is weighted in a manner correlated with wealth. The median influencer in these economies is not a billionaire, but an older citizen of some affluence who has mostly endowed her own future consumption. She would like to be richer, of course. But she is content with her present wealth, and is panicked by the prospect of becoming poorer. For such a person, the depression status quo is unfortunate but tolerable. The risks associated with expansionary policy, on the other hand, are absolutely terrifying.

The revealed preference of my polity is not my personal preference. Perhaps that is because I’m an idealist, and I actually care about the misery provoked by precarity and unemployment. Perhaps it’s simply because I’ve not yet endowed my own future consumption, and I’m scared. Regardless, I object. Although I understand where it comes from, I detest the preference for depression revealed by my polity. Perhaps you do too.

But if we want to change the behavior of the polity, it’s not enough to argue over clever policies that, if implemented, might do the trick. We’ve got to change its preferences, which means either buying off the median influencer, or changing her identity via political struggle. Alternatively, we can wait until what are now problems of aggregate demand morph into supply problems (after people become unemployable and capital decays), or into threats of political and social unrest. The median influencer may change her views if tight supply makes goods costly despite fiscomonetary conservatism. Or if her neighborhood is on fire. But I’d prefer we avoid all that, and take a more proactive route.

In the meantime, we have to recognize that what we are experiencing is not a technical failure. It is not “magneto trouble”. We, collectively, are making a choice. The task before us is to change our mind.

Update History:

  • 17-Apr-2012, 9:20 p.m. EDT: Small edits to eliminate wordiness and word repetition: “In the meantime, what we have to recognize is that what we are experiencing is not a technical failure.”; “and is terrified frightened of becoming poorer”
  • 17-Apr-2012, 11:50 p.m. EDT: Changed my change above; “frightened” is too weak: “and is frightened panicked by the prospect of becoming poorer”
 
 

247 Responses to “Depression is a choice”

  1. Is your analysis of Australia:
    (1) It has a different revealed preference
    (2) It would have the same revealed preference in the above situation, but its policies mean it never got there in the first place?

  2. Olympia Carraway writes:

    Great publish it can be truly. We’ve been awaiting for this info.

  3. Dan Kervick writes:

    I have never understood Yglesias’s claim about cash, the zero bound and negative interest rates. I really don’t think he has thought through the mechanisms very carefully, and is putting way too much emphasis on the role of physical currency. Make all the money electronic and nothing changes. There is no circumstance under which a commercial bank has an interest in making a negative interest loan, since that involves the exchange of an asset for an asset of lower value.

    And like many, he gets the role of interest on reserves exactly backward since he thinks of banks as lending their reserves.

  4. Foppe writes:

    Well, it can be a choice for depression only if you take the system to be the relevant entity of analysis. But it seems to me more informative to look at smaller collectives.
    In that light it is probably useful to remember that the state apparatus is used quite frequently to nip attempts at collective action in the bud, so to say (cue certain pictures of pepper-sprayed students and movies of veterans being beaten to a pulp by Oakland PD police officers here). But it is not a new insight that successful protest actions generally require a lot of preparation and previous protests that went nowhere; so the more successful governments are at discrediting and dispersing previous attempts, the longer they stay in power.

  5. Agog writes:

    IMO a big part of the problem is that many of the current cohort of technocrats have spent their whole careers learning to solve the problems of the 1970s. And trying to prove, time and time again, that yes they have truly learned how to solve the problems to the 1970s…

  6. m64 writes:

    I think this is a very valuable insight. I so far have only worked for about seven years, but I can already tell that my perspective is much different from that of my younger relatives who are just entering the workforce. I am quite well off and can just sit out the recession without much trouble, even if it were to last for another 5 years. Hell, I could probably live for a year just out of my savings account if prices stay stable. I don’t exactly need a revolution. For them another 5 years without job security can be a life breaking tragedy. And I think this will only get more and more like that as I get older and acquire more savings and property. The risk is not in the interest of the median influencer.

  7. K writes:

    Dan: If banks can fund in the interbank market at a negative rate, then lending at any rate of return above that rate is profitable at the margin. Period.

    I don’t understand what you mean by “that involves the exchange of an asset for an asset of lower value.” The lower value asset is 1) in the future and 2) of lower *nominal* value. There simply is no risk-free way to transport a fixed nominal quantity into the future so they have no choice but to take a nominal loss. But I assume you are talking about *real* value. Banks already make loans at lower real value when they buy T-Bills at a real return of -2%. Banks don’t care about the real or nominal return of their assets. What they care about is the risk adjusted return of asset *plus funding*. Nothing changes below zero. Yglesias is exactly right.

    Steve: Beautiful piece. I love the idea of depression as revealed preference. But I’m not sure if you are right that eliminating the zero bound wouldn’t make any difference. There is an irrational fear of inflation brought on by massive QE, which goes along with the irrational belief in QE in the first place. If we could have just cut rates to -5% or -10% in the first place instead of this ridiculous fraud, we would never have got into this trouble in the first place. Also, the Fed could counteract overexpansions much more aggressively without fear of suddenly transitioning to intractable recession. In general, without fear of the liquidity trap, monetary policy could be much more aggressive on both sides.

    I’ll admit a worry though: Lets say the Fed cuts to -5% and pumps the economy back up through consumer borrowing. Then with tons more debt, we are all the more fragile against the next shock, down go rates, and on and on turns the ratchet. What is the asymptotic leverage equilibrium? How about volatility? Where does the Great Moderation end?

  8. I think is more subconscious than anything. The banks are under water, ask for help and the people vote not to help them anymore. So rates go up while loan approvals go down and no one can afford anything. Its like a Benny Hill skit chasing eachother through the hallways.

  9. bryan willman writes:

    the behavoiral problem is deeper than the storing of money in preference to consumption. it is a problem about demographics, about limited needs, and “enough”

    so inflation or negative interest are used to drive down the value of holdings – it is assumed that this will drive up consumption. but most of the polity holding on in fear are not rich – so they will just cut back even more to stay within their means. the largest credits are retirement claims, soc sec benefits, ordinary personal savings, and earnings from jobs so long as people are employed. reducing the real value of any of those things will never be stimulative.

  10. Dan Kervick writes:

    Dan: If banks can fund in the interbank market at a negative rate, then lending at any rate of return above that rate is profitable at the margin. Period.

    K, if I am a bank and make a loan for $10,000 at a negative rate of interest, then I create a demand deposit for the loan customer for $10,000, and exchange it for an IOU from the customer for some amount less than $10,000 to be paid at a later date. The demand deposit is a liability for $10,000 and the IOU is an asset worth less than $10,000. It is simply not in my interest to make that swap. And the calculation is the same if instead of creating a demand deposit for the customer I instead give the customer $10,000 of vault cash.

    This is true no matter what the cost of reserves, and even if the central bank is charging negative interest for the reserves – i.e., paying me to borrow them. That’s because borrowing the reserves at negative interest and not making a negative interest loan turns out to be more profitable than borrowing the reserves at negative interest and making a negative interest loan. If the central bank engineers a policy rate that allows banks to increase their net assets simply by borrowing reserves, then why in the world should they diminish those assets by generating loans that are a net liability?

    Inflation rates and the distinction between real and nominal value do not affect this calculation at all. The loan contracts are written in nominal terms. If I have $10,000 now and can have that $10,000 one year from now by not lending it, that will always be better for me than having only, say, $9500 one year from now. If inflation is eating away at the value of money, so that my $10,000 will be worth much less one year from now, then it will still be the case that the real value of $9500 one year from now is only 95% of the real value of $10,000 one year from now.

    Yes, if the cost of my funds for the period of the loan is -5% and I make a loan for -4%, then the loan is still profitable for me to the tune of 1% over my costs. But what is even more profitable for me is to borrow the funds at -5% and then not to make the -4% loan. Then my profit is 5%.

    There is a risk-free way to transport a fixed nominal quantity into the future. Hold onto it. It’s value might decline; but it will not be any less than the value of some lower nominal quantity.

  11. Steve Roth writes:

    As usual, top notch. We as a polity, in which those who have endowed their future consumption (de ebil creditors) dominate both voting and campaign finance, have ceded control of the monetary system to the Fed. This is congenial to that dominant block because the Fed is run by creditors.

    In answer to the question of why incumbents wouldn’t want strong growth, Ryan asked and I answered:

    “in what way is higher inflation different in its impact on the elite than negative interest rates?”

    Inflation (or at least unexpected inflation [though one could argue that all inflation is unexpected; cf experts' current predictions of both inflation and deflation]) instantly and permanently deflates the real value of the massive *stock* of creditors’ holding. We’re talking: one percent extra inflation, hundreds of billions of dollars in real buying power (for things humans can consume, including long-term goods) transfered from creditors to debtors. From holders of financial assets (especially bonds but also stocks) to holders of real assets (including the ability to work).

    This effect utterly swamps the flow effects resulting from low interest rates.

    And: creditors are myopic. Short term, declines in interest rates give them cap gains on their bond holdings. Add that to the short-term hit from higher inflation, and the Fed’s incentives (remember, it’s run by creditors) are pretty clear.

    I wonder:

    Suppose the Fed went to negative interest rates, but refused to provide additional physical currency (though maybe they’re required to?). People could still spend through all the electronic payment systems.

    Might result in a depression for the poor, who don’t have access to those mechanisms. But everyone with debit and credit cards and paypal would be okay, no?

  12. Tom Hickey writes:

    Agree. Traders know that politics and policy choices are major factors in market behavior.

  13. Steve Roth writes:

    @bryan william: “so inflation or negative interest are used to drive down the value of holdings – it is assumed that this will drive up consumption. but most of the polity holding on in fear are not rich – so they will just cut back even more to stay within their means. the largest credits are retirement claims, soc sec benefits, ordinary personal savings, and earnings from jobs so long as people are employed. reducing the real value of any of those things will never be stimulative.”

    In the short term, declining interest rates *increase* bond-holders’ nominal holdings. They watch their brokerage balances rise. Yeah, they’re worried about rollover in the future, but now is…*now.*

    Increased inflation has no impact on their nominal balances, but they are watching their stock of real wealth evaporate.

    Myopia means they want low inflation and are willing to tolerate low and declining rates for a long time because they’re seeing (perhaps long-term illusory) cap gains.

    But what they really want is low inflation and high interest rates, no matter its effect on the macroeconomy.

    The Paradox of Wealth?

  14. [...] Steve Depression is a choice! We only thought that preferences were over growth and employment but [...]

  15. Dan Kervick writes:

    Perhaps it is worth pointing out that paying interest in reserves is one way in which the Fed is already moving toward negative rates. IOR is a zero bound tool. If a bank has to pay X% interest to borrow reserves for some period of time, but the fed is paying the bank a positive rate of interest on those reserves during the period of time they possess them, then the Fed is effectively giving the bank a discount on their borrowing cost. By increasing the IOR up above the Fed Funds rate, the Fed would be doing something that is functionally the same as having a negative Fed Funds rate.

  16. K writes:

    Dan,

    “Hold onto it.”

    Hold on to *what*? You’re thinking like we’re passing around some kind of finite physical commodity. There’s no “it” (unless you are talking about some regulated quantity of reserves – but those earn negative interest).

    “But what is even more profitable for me is to borrow the funds at -5% and then not to make the -4% loan.”

    Borrow the funds at -5% and do *what* with them? You have to pay the money back, so you’ll have to hold *something* of value until you repay that loan.  The debt isn’t going to repay itself. What, exactly, are you going to hold?

    Walk me through the exact balance sheet operations that you are envisioning.

    “inflation rates and the distinction between real and nominal value do not affect this calculation at all.”

    Agreed. I thought maybe that was what was causing the confusion. Just ignore it.

    Steve,

    Actually let me put it stonger: QE was a fail, and ought to have been expected to be a fail. There is absolutely no good theory that predicts that it would work when the assets are negative beta treasuries (even if they are stocks, you have to include your model, exactly which economic agents indirectly take on the purchased assets via Ricardian equivalence – not obvious, and no such model exists`). The quantity of policy rate-bearing reserves is, of course, irrelevant.

    But because lots of people thought it might work, it sabotaged effective fiscal action. But negative rates *would* have worked without ever having to go fiscal at all. That’s why we would have been way better off in a cashless economy (at least for the time being).

  17. Dan Kervick writes:

    As I wrote in a comment above, no matter how low the Fed Fund rate goes, even into negative territory, that in itself will never bring it about that commercial banks have an interest in offering negative nominal rate loans. The problem Yglesias is concerned with – that holding onto monetary assets for some period of time is always better than trading those assets for a smaller nominal quantity of the same assets at the end of the same period of time – applies just as well to banks and their reserve assets as individuals and their physical currency assets. If you want to get actual negative interest lending going to households and businesses, you will need more far-reaching changes.

    One approach would be to limit the ability of banks to borrow additional reserves except in cases where they have written additional loans. This strikes me as a huge bureaucratic and regulatory challenge, and a large change in the way banks organize their businesses and the way the Fed Funds market functions, but maybe it could be accomplished. The Fed could change the system and time frames for calculating reserve requirements and require banks to present some accounting of their total reservable deposits in order to borrow additional reserves. They can only borrow more reserves if the deposits have gone up. Maybe they present the accounting, the Fed declares, “OK, because your deposits have gone up by $10 million and you were already right at the required reserve ratio in the previous period, you are entitled to borrow $1 million in additional reserves.” They issue a $1 million electronic “permission slip” to the bank which it then takes into the Fed Funds market to borrow.

    Another more radical approach would be to move to a system of public banking – or at least a public option in banking. Public banks could have operating policies that are based on the aim to fulfill public purposes, not make a profit. There are no reserves in the public system. Each bank is basically just a sort of loan desk for the Fed. The Fed sets the lending rate and that’s it. They can set it at some negative amount if wished. Bank officials and loan officers are evaluated not in terms of the amount of money they make for the bank, but in terms of the amount of the loaned amounts that are recovered from the public.

  18. Dan Kervick writes:

    Borrow the funds at -5% and do *what* with them? You have to pay the money back, so you’ll have to hold *something* of value until you repay that loan. The debt isn’t going to repay itself. What, exactly, are you going to hold?

    You hold the balance itself in your reserve account. Assume the Fed Funds rate is -1% for an overnight loan. If you borrow $1,000,000 today, you will owe $990,000 tomorrow. So you borrow $1,000,000. $1,000,000 is credited to your reserve account. You then go home and play Words With friends all night. The next day, you pay the lender the $990,000 you owe him. Your debt is discharged. Net result: Your reserve account is $10,000 fatter; you made $10,000 for doing nothing.

    Now you could reason, “If I make an overnight loan to my brother-in-law for $1,000,000 at -2.5% interest, then tomorrow he will pay me $995,000 and I will still have netted $5,000.” That makes sense if you are a nice guy. But if you are a profit maximizing bank, playing Words with Friends and making no loans as you sit on your magically reproducing reserve balance is more profitable than loaning money to your brother-in-law at negative interest.

  19. K writes:

    Dan: “By increasing the IOR up above the Fed Funds rate, the Fed would be doing something that is functionally the same as having a negative Fed Funds rate.”

    They can’t do that. You can’t have a fed funds market if IOR is above the fed funds rate. Why would anybody lend FF if they can hold reserves at a better rate? If there’s no fed funds market, there’s no fed funds rate.

    Unless you are at the mandatory reserve boundary, reserves must earn essentially the same rate (with minor technical details) as the interbank rate. In countries that have abandoned mandatory reserves and in the countries that have increased reserves above the mandatory quantity via QE, the reserve rate is essentially equal to the policy rate.

    Canada, for example, has paid the policy rate on reserves for a long time, whether that rate was 1% or 7%. It just underscores the fact that the BoC implements policy via control over (real) lending rates, not via some monetary “hot potato.” That doesn’t mean Canadians are any closer to getting negative rates than anybody else. This might though.

    Still reading your subsequent comments…

  20. K writes:

    Oops. Forgot link.

  21. Dan Kervick writes:

    Can I also suggest that a negative interest rate regime, if it could be implemented, is just another term for a debt jubilee? Everybody can borrow a quantity such that the difference between the principle borrowed and the total repayment is exactly the amount needed to pay all of their existing debts. All creditors get paid.

  22. Dan Kervick writes:

    They can’t do that. You can’t have a fed funds market if IOR is above the fed funds rate. Why would anybody lend FF if they can hold reserves at a better rate? If there’s no fed funds market, there’s no fed funds rate.

    Good point K. So that seems to suggest that you can’t have a negative Fed Funds rate without negative interest (a tax) on reserves, right? The only reason banks would be willing to lend reserves to other banks at negative interest would be if the quantity of those reserves was being drained even faster by not lending them at all.

    But all of this is on the liquidity side of the bank’s accounting. No matter what is happening on that side, and no matter how bad it is for them due to CB policies, they only make their balance sheet worse by making commercial loans at negative interest.

  23. Woj writes:

    Great insight as always. I agree on the trading/investor side. With little political chance of seeing much higher inflation anytime soon, even at current rates, bonds offer reasonable value.

  24. Nick Rowe writes:

    Steve: good post.

    One tiny change: “The purchasing power of holders of [safe] nominal debt must not be put at risk.”

    And the coalition in favour of depression is broader than that. I have had people argue against me, that increasing NGDP growth would be a bad thing for workers, because it would reduce real wages!!

    Maybe, it’s because most people never think past step one.

  25. himaginary writes:

    So the depression is the “Id Monster” in Forbidden Planet, so to speak. Our financial system is now working as the Krell Great Machine.

  26. Steve Waldman discovers end-game political economics is a Darwinian struggle with no winners only those who are prepared to lose less than others.

    Steve Waldman assumes that Americans and others can simply choose to live beyond their means as we have for decades.

    Steve Waldman assumes like thousands of other ‘Brand X’ economists (Keynes) that the industrial waste-based economy is ‘productive’. It’s not …

    :)

    Their overwhelming priority is to protect the purchasing power of incumbent creditors. That’s it.

    The incumbent creditors are THE creditors! There are no other creditors.

    A tiny problem is that industrial enterprise does not pay for itself. Except for taxi drivers and deliverymen, driving a car and all the trillion$ invested in the experience around the world does not produce a return. It is pure waste for its own sake that must be propped up with bottomless debt- and energy subsidies. This is the reason for the tens- of trillions of dollars worth of debt taken on in the first place. If industry could pay for itself it would have! There would be no debt because returns earned by industry would have retired it!

    Two dynamics are spooling out:

    – where repayment requirement exists on the margin,

    – the effect of very large numbers.

    With a multi-decade worldwide increase in credit on largest scale the issue arises about repayment which is the difference between the ‘loan’ concept and the ‘gift’ version. The creditors are seeking retirement of the tiniest amount of finance debt (rather than debt roll-over or re-financing). Repayment means repayment: it does NOT mean default/failure to repay leading to bankruptcy. Repayment at some margin is essential otherwise the concept of ‘credit’ collapses under its own weight (which is taking place right now, under everyone’s noses).

    See: ‘Greece’ …

    Finance debts are too large to repay. They can only be rolled over endlessly or repudiated: There is No Other Way! Adding to the amount of debt is pointless as the costs increase while the returns on the debts themselves (productivity of debt) declines.

    Who is to repay? (Who can?)

    The treasury (any treasury) can issue pure fiat (on its own account w/o liability) and use the currency to retire debt as it comes due. Creditors howl because their debts are expected to represent claims on real goods and services rather than abstractions that are no different from the debts, themselves. This is a form of repudiation.

    The state taking private agents’ assets and liabilities (both) onto its own balance sheet and extinguishing both is also a (conventional) form of repudiation. The state can repudiate outright dead-money and onerous debts (and prosecute/execute those who issue such ‘loans’).

    Executing the marginal creditor might appear to be counterproductive but acquiescing to every creditor demand (then declaring bankruptcy) has proven to fail. Selective repudiation at the margin creates an incentive to compete to be ‘survivor’.

    Central banks cannot create new credit, only recycle what exists (starting with that which is impaired). Central banks are collateral constrained (otherwise they are not central banks).

    The issue shifts from credit that has run aground on its own internal contradictions toward what cash money (currency) is worth when priced in credit. This is also a fail because the money cost of money is irrelevant …

    The worth of money is determined by millions of motorists around the world exchanging money every day for a valuable natural resource/good on demand. The ‘competition’ is between worth-determining agents. Even with negative interest rates administrators are unable to fix the money-price of money. We now have de-facto hard currencies consequent to 200+ years of ongoing capital waste. Any attempt to encourage more credit/money creation (on the part of those fore-mentioned creditors) pushes the price of crude to unaffordable levels. The price of fuel is too expensive: wasting fuel costs more than it returns. Our ‘debtonomy’ is run aground on the inability of waste to pay for itself (and now to even service its own debts).

    The worth of money in 2012 relates directly to the petroleum that can be exchanged for it. Be thankful the worth of money isn’t the worth of drinking water. When that day comes you will wish you are dead (you probably will be dead).

    The money hoarded represents real oil/gas and diesel. People expect to sell their gas and diesel hoards quite dear when the time comes. As JP Morgan said when asked about America’s Cup yacht racing, “If you have to ask you cannot afford it”.

    Welcome to the age of rational expectations!

    This dynamic — not administrative ‘choice’ is the reason for our ‘depression’. I hope you like it, it is going to get a lot ‘worse’. There is nothing anyone can do to prevent worse from taking place, either. Deep in y’alls hearts you know I’m right.

    :)

    BTW: Steve, yr brilliant, keep up the great work!

  27. Vinz Klortho writes:

    I strongly recommend the reading of “Secrets of the Temple – How the FED controls the economy” by William Grieder. It recounts in excruciating detail the actions of the FED during the inflationary 70′s and 80′s, and the high interest rate/strong dollar economy destroying actions they took during those times. The FED will, if necessary, destroy the debtors and credit dependent entities in the economy to protect the creditors against inflation. Paul Volker was lobbied heavily by govt representatives from the farm, manufacturing, credit and export dependent states, and told them “Gentlemen, your constituents are unhappy, mine aren’t”.

    I read this ten years ago, and got a lot out of it. Reread it this last month, with the benefit of a lot more knowledge of banks, FED and financial system (thanks all bloggers!) and got A LOT more out of it.

    Vinz

  28. K writes:

    “So that seems to suggest that you can’t have a negative Fed Funds rate without negative interest (a tax) on reserves, right?”

    Exactly. Reserves must *always* make less than (or the same as) the policy rate.

    “they only make their balance sheet worse by making commercial loans at negative interest.”

    Not so! They borrow at -5% and lend at -4%. The only thing that matters, whether you are above or below zero, is the difference between the lending rate and the funding cost. There is literally no impact of going negative *other* than the critical fact that the *real* policy rate can go below negative the rate of inflation.

  29. [...] Steve Randy Waldman, “We are in a depression, but not because we don’t know how to remedy the problem. We are in a depression because it is our revealed preference, as a polity, not to remedy the problem. We are choosing continued depression because we prefer it to the alternatives.”  (Interfluidity) [...]

  30. EmmaZahn writes:

    Very important and well articulated point.

    Thank you.

  31. You are a genius, and you are absolutely correct. Thank you for existing and writing.

  32. Dan Kervick writes:

    Not so! They borrow at -5% and lend at -4%. The only thing that matters, whether you are above or below zero, is the difference between the lending rate and the funding cost.

    I wish I could draw a decision matrix here, K, because I think the point would be easy to make.

    Imagine across the top line of the decision matrix two possible Fed policy choices: (I) Fed Funds rate at -5% and (II) Fed Funds rate at +1%.

    Now imagine on the left-hand side of the decision matrix these three possible bank decisions: (a) borrow $1 million at the Fed Funds rate and make a $10 million loan at -4% interest; (b) borrow $1 million at the Fed Funds rate and make no loans; (c) borrow $1 million at the Fed Funds rate and make a $10 million loan at +4% interest.

    So there are six possible outcomes with six different bank payoffs. If you calculate out the numbers, you will see that (b) is superior to (a) for the bank, no matter which Fed Funds rate the Fed sets. And it is also true that (c) is superior to both (a) and (b), again no matter which Fed Funds rate the Fed sets. (b) dominates over (a), and (c) dominates over both (a) and (b).

    Now, one might point out that with the Fed Funds rate set at a negative number, (c) might not be an option for the bank. Competitive pressures in the commercial banking industry will drive the commercial lending rate down to just above zero. But those pressures can never drive that rate down below zero. Because not lending at all is always/i> an option for the bank no matter what the Fed does, and (b) is always more profitable than (a). If your competitor says, “Hey, I just made a -4% loan of $10 million by borrowing the needed $1 million reserves at -5% interest, and I made a $10,000 profit!”, your response should be, “Idiot. I just borrowed $1 million in reserves at -5% interest, made no loans at all, and made a $50,000 profit!”

    What you are pointing out is that outcome (Ia) is profitable for the bank. But what I am pointing out is that outcome (Ib) is even more profitable for the bank. And since (IIb) is also more profitable for the bank than (IIa), then (b) is more profitable than (a) no matter how you slice it. As I said, I am not disputing that making the loan at a negative rate of interest is profitable if the cost of funds is negative. What I am pointing out is that it is not as profitable as borrowing the reserves at the profitable negative interest rate and making no loan at all.

  33. Dan Kervick writes:

    Sorry about my tags.

  34. Benjamin Cole writes:

    Excellent blogging. Keep up the good work.

  35. beowulf writes:

    “They can’t do that. You can’t have a fed funds market if IOR is above the fed funds rate. Why would anybody lend FF if they can hold reserves at a better rate? If there’s no fed funds market, there’s no fed funds rate.”

    Fed Funds rate averaged 0.15% yesterday, IOR rate is 0.25%. Two reasons for the differential, 1. GSEs aren’t paid IOR, so they’ve bid down FFR, and 2. thee FDIC’s new assessment system (Dodd-Frank changed the deposit insurance “tax base” last spring so its now a de facto tax on reserves). As was pointed out in Sept FOMC meeting…
    “a recent change in deposit insurance assessments had the effect of significantly reducing the net return that many banks receive from holding reserve balances.”
    http://newmonetarism.blogspot.com/2011/10/twisting.html

  36. Truth Squad writes:

    We live in oligarchies, not democracies. In oligarchies the interests of the oligarchs are protected and they have the control, if needed by manufacturing ideologies and promoting them, and by using political power.

    We haven’t decided anything. No one in Europe has been asked about what they want to do, elites tremble sometime someone mentions the ‘referendum’ word, no one in USA was asked about if they wanted the TARP either, I could go on.

    The notion that the people has decided anything is foolish and only reinforces the status quo: that you have a choice, it’s your fault the situation you are in, you decided it. Sorry, but I don’t buy that bullshit, I haven’t decided anything, neither has the majority of the people. See on how the political machine works, how it demotes or promotes political leaders, etc. in every single nation of this planet (each one with its own idiosyncrasies). And let’s not get started about the part of promoting wrong ideas and using propaganda to push hidden agendas.

    Very few people have decided the fate of whole nations and humanity, they asked no one, and they will continue to do so as long as they can. You don’t have a choice, there is no choice. Maybe at some point situation will get so awful that circumstances will force compromises, but that moment has yet to come and as it’s seen (Greece), is actually not that easy to get there. Specially when there is such strong capture of a part of the population that will do ANYTHING to keep their situation 8that’s how authoritarian governments rise, in the shoulders of desperate ‘middle classes’).

    And no, there is no conspiracy theory, it’s something anyone with a brain and some observation can testimony day after day of watching all this situation develop. Until you acknowledge this, there is no point at talking about policies or economic theory, the dominant theories and policies will be the one out of the consensus of the different TPTB.

  37. Truth Squad writes:

    Ok, a -1 for my comment above, I should have actually continued reading, as he more or less points this problem.

    In short: our fate is in the end in the hands of our complacency, and that’s what will destroy (it already is in some countries) us. In the end we will have civil wars, revolutions and maybe even a World War. I don’t see an other way around it, if history is of any guide, it has been this way since the first civilizations.

    Stupidity and complacency triumphs.

  38. Becky Hargrove writes:

    You could put up a new post every day and we would all be happy. I will just contribute a scribble from earlier this morning: The poor don’t have any more time to wait for the upper and middle classes to squabble over the way consumption baskets are currently constructed. They will have to find their own solutions and ways to survive, without having to constantly rely on others who are already overwhelmed.

  39. K writes:

    Dan,

    “I just borrowed $1 million in reserves at -5% interest, made no loans at all, and made a $50,000 profit!”

    One more time: What did you do with the money you borrowed??? *How* did you carry your $1 million forward in time and still have $1m at the end? THERE IS NO NOMINAL INSTRUMENT THAT EARNS ZERO PERCENT!!! Again it’s like you think money is some kind physical commodity that, at the end of the day, you can just hold. Well you can’t, Yglesias revoked it. I like talking to you, Dan, and you have a lot of insight to contribute. But on this topic I feel like you’re trapped in some kind of extreme money illusion.

    There is no cash, man!!!

    As far as the rest of your example goes, I’m assuming that deposits make up the missing $9 million of funding and that deposits earn something in the vicinity of reserves/FF? With competitive deposit and repo markets, all these rates would be the same.

  40. Dan Kervick writes:

    One more time: What did you do with the money you borrowed??? *How* did you carry your $1 million forward in time and still have $1m at the end?

    K, you don’t need physical cash. The bank’s money exists as an electronic balance in its reserve account at the Fed. That electronic balance already is money. It’s part of MB. It’s part of the bank’s assets and total reserves. Physical currency is irrelevant. When the bank borrows the $1 million on Monday, that $1 million gets credited to its reserve account. When it pays back $950,000 on Tuesday to repay the loan, that $950,000 gets debited from your reserve account. That’s it; all done. The bank now has $50,000 more is assets than it had before.

    The way the bank carries forward the nominal assets in its reserve account from time A to time B is simply by not making a withdrawal or deposit to that reserve account from time A to time B. It doesn’t matter whether the bank ever orders physical currency from the Fed. It all ready has the money in electronic form. If the Fed is paying interest on reserves, then even better for the bank. Not only is it able to carry forward its reserve balance; that balance actually increases even if the bank does nothing.

    What other $9 in funding are you talking about? A bank does not have to collect deposits first to make a loan. If a bank wants to make a loan for $10 million, it makes a loan for $10 million. It credits a loan customer’s account by $10 million. If it was already carrying excess reserves, then it might not need to acquire any additional reserves at all. If it does need to acquire additional reserves, then it acquires them afterward. And it doesn’t need to acquire $9 million in reserves – in the US system it would just have to acquire $1 million. It might acquire those additional reserves through additional deposits, but it could acquire the whole $1 million by borrowing them. There is no other $9 million that the bank has to come up with somewhere.

    This is the same business Scott Fullwiler went through with Krugman, which Krugman didn’t seem understand initially. A bank does not make its loans out of a stock of funds that have been put on deposit. Loans create deposits, which in some cases require additional reserves, but in some case does not.

  41. Dan Kervick writes:

    SRW,

    Your post is great, and I neglected to say that. Sorry to have clogged up the comments section with the side-discussion about Yglesias on cash and negative rates.

  42. Oliver writes:

    The interests of the median influencer are aligned with those of the working population to the extent that a bigger pie is a bigger pie. The problem lies in trying to guarantee a piece of the fictive pie over many years via interest and inflation rates so that the theoretically congruent, or at least potentially negotiable interests between pensioners and workers become realigned into a broad coalition of what you call a cartoonish capitalist-class on the one side and the working classes on the other. The capitalist-class is heterogenous, perhaps merely an aggregation of parts of every one of us, but it is a real economic entity nonetheless. And while I quite liked your proposal for starter saving accounts, it doesn’t deal with the problem that the large pools of superannuation and private pension fund wealth pose to the financial system. What would the savings from the starter accounts be invested in?

  43. K writes:

    Dan: “The bank’s money exists as an electronic balance in its reserve account at the Fed”

    OK, but I thought we agreed the reserve account earns negative interest, at a rate equal to or less than the FF rate. No? So on Tuesday, there’s only $950K in the reserve account, right?

    “What other $9 in funding are you talking about? A bank does not have to collect deposits first to make a loan.”

    Banks don’t have to check if they have enough money to lend you, as you say. But they still have to fund the loan, right? Assets=liabilities and all that? Lets say I write you a $10 million check on my line of credit. On the day that the check clears, my bank enters on its balance sheet a loan to me of $10 million and sends a Fed Funds transfer to your bank. Your bank credits your deposit account with $10m. This all happens at once: my bank lends some money and it ends up in your account. So when my bank sets the overnight lending rate on my line of credit, it doesn’t assume that my negative balance will be funded by a corresponding positive deposit balance. They *can’t* assume that. Deposits that result from lending are quickly dispersed through the system. So, before the end of that same day, my bank has to *borrow* $10m from your bank in the interbank market in order to fund the Fed Funds payment. The deposits don’t automatically come back by magic.

    So there is *no* operational link between the decision to lend, and the arrival of funding for that loan on my banks balance sheet. When they agree to lend to me, they assume that they will be funding that loan from a mix of sources that depends on the riskiness of the loan. If it’s a *very* low risk loan (i.e. a loan to a government) then little or no additional capital will be required, and the loan can be funded entirely from the interbank market (pari passu to deposits). Either way, the interbank market sets the base rate for the banks funding costs. To the extent that additional capital is required, the bank will add the costs of those kinds of capital to the loan rate.

    “There is no other $9 million that the bank has to come up with somewhere.”

    Yes, there is!

  44. K writes:

    I meant “sends a Fedwire transfer to your bank”

  45. Floccina writes:

    Another reason that it was a bad idea to leave the monetary system, to the median voter. So not only are they rationally ignorant about the monetary system they might even rationally be motivated to prefer unemployment to some reasonable level inflation.
    Free banking now!

  46. W.C. Varones writes:

    You people are nuts.

    Ban cash and turn bank savings into a wasting asset?

    Watch the savers and investors take their capital where it’s wanted: to a free country far away from your central planners’ paradise.

  47. Greg writes:

    Excellent post Steve. One of your best.

    Its actually pretty well known (might could even be called a fact by now) that we are much more loss averse than we are gain favorable so none of this should be a surprise. And of course those with the most to lose are the MOST loss averse. Having much can probably be said to make you stupid. You will go to very irrational lengths to cut your losses.

    Its also true that we dont like seeing others of our kind suffer, and this might offset the forces of loss aversion………. but we’ve developed a pretty affective means of diminishing the suffering of others (in our minds). 1) Dont show it. Our media very rarely shows the true level of hurt in this country 2) When its shown, diminish it somehow. FInd a way to dislke the people. If they are suffering and screaming indignantly at the camera its better than if they just are silently suffering. 3) if the sufferers actually own something like a car or a cell phone, its easy to say they would suffer less if they didnt own those things, so its their CHOICE to be where they are.

  48. Lord writes:

    You can view this as a psychological or even technological explanation. We believed the internet and then housing would make us all wealthy. We believed in boundless opportunity and endless growth. We believed in the future. We no longer do. No longer do we have new ideas. No longer are investments worth making. Our possibilities have become constrained. Our horizons have contracted. Our faith shattered. We recede into preservation, eking out our lives, getting by and making do with what we have.

  49. Morgan Warstler writes:

    Steve, oh come on man, let’s REALLY stare into the abyss.

    Money, monetary policy, and govt. itself is not constructed from thin air, it has real knowable DNA. Needs for certain people it was made to meet.

    Nowadays it is the owners of property. They guys who WANTED to hire big guys and have them protect their stuff. To want that, you gotta have: 1) stuff 2) enough extra stuff to pay big guys.

    That’s govt. So don’t be surprised if the monetary system isn’t neutral to the concerns of Steve and those that aren’t of the creditor class. That’s a FEATURE.

    It isn’t a “preference” – even your word choice acts as though the entire structure isn’t a monolith against what you want.

    You see progressives make the same mistake about US law. Our founding document is BUILT on making sure the local kingpins in each state, don’t have to worry about the national kingpins taking over.

    BUT STILL, they go things done.

    And how did they get it done??? By letting the other side, the other state king pins have what they wanted – no worry of losing their status.

    Ok, you want to move the economy forward – we have 30M people who could be working.

    So we give them a Guaranteed Income Paypal style ($240 per week), and we auction the unemployed Ebay style (starting at $40 per week).

    THIS IS THE SOLUTION. This is the one the POLITY will give you Steve, if you just ask.

    Everyone gets a livable income.
    Everyone get a private boss pushing them 40 hours a week.

    The real reason it is the solution is because no one who matters loses status they currently have.

    Now the govt. isn’t trying to direct employment.
    Businesses have the downward push on wages.
    Unemployment disappears, there is ZERO excess capacity.
    Goods and services for the lower class get far, far cheaper.

    In effect Federal Spending is used to make everyone comfortable, but LIBERALS ADMIT the unemployed are best used by the private sector (who pay the govt. bills) seeking profits for themselves.

    You can have half a loaf. Stop daydreaming about revolucion.

  50. Dan Kervick writes:

    OK, but I thought we agreed the reserve account earns negative interest, at a rate equal to or less than the FF rate. No? So on Tuesday, there’s only $950K in the reserve account, right?

    I thought you and I were talking primarily about a negative Fed Funds rate K, so that as you said earlier “banks can fund in the interbank market at a negative rate” and then loan commercially at a negative rate above that. If the Fed taxes reserves (sets negative interest) at a daily rate functionally near or equal to the Fed Funds overnight rate, then the two policies would offset. The overall cost of funds would no longer be negative, but somewhere closer to 0%.

    Suppose the FF rate is 2% and IOR reserves is 2%. This is similar to having a 0% rate of interest with no IOR. The lending bank loses interest income from the Fed from each dollar it lends, and that loss offsets the interest it receives from the borrowing bank. The borrowing bank gains interest from the Fed that offsets each dollar it has to pay to the lending bank. The same is true if the FF rate is -5% and IOR is -5%. The reserves you acquire cost money to hold, but that is offset by the fact that the lender is paying you the same amount to acquire them. If you want the actual cost of acquiring reserves to be -5%, then it seems to me you either want a -5% Fed Funds rate with no interest on reserves, or you want all reserves to be borrowed directly from the Fed at 0% interest, but with the Fed paying 5% interest on reserves.

    You are right about the line of credit scenario. But most deposit balances are not withdrawn or used as payment during a given period at anywhere close to the total size of the deposit. That’s why reserves are permitted to be so low in relation to deposits. The bank’s ratio of reserves to deposits is somewhere around the reserve ratio. The assets that the bank acquires to balance its liabilities are the IOUs from the borrower to repay the loan with interest, not its reserves.

    Suppose a hypothetical bank had only a single customer deposit account, created by a loan, in the amount of $1,000,000. It will then have to carry total reserves around $100,000. The deposit is a liability and the reserves are an asset. So considering only these two items, the bank’s balance sheet shows a net liability of $900,000. But the bank didn’t make the loan for nothing. It exchanged the deposit balance for an IOU from the loan customer. The nominal value of the IOU will be for $1 million plus some interest, and the present value of that IOU, while not as great as the nominal value, will (hopefully for the bank) still be somewhere in excess of the value of the deposit balance. That’s how its assets exceed its liabilities. A bank doesn’t fund its loans with stocks of pre-existing money. It funds its loans with promises, and broad money comes into existence as the overall social stock of promises expands.

    So, there is no additional $900,000 in our original example. If you are already carrying $100,000 in extra reserves, you can make a loan for $1,000,000. If you don’t have any excess reserves, you can extend a loan for a million dollars and then turn around and borrow the needed additional $100,000 in reserves at the going FF rate. You charge your customer some interest which means they owe you in return $1,000,000 plus change. And you only need to have enough reserves during any given period to accommodate the actual cash withdrawals from and payment orders written on your total deposits.

  51. K writes:

    Dan,

    “So, there is no additional $900,000 in our original example. If you are already carrying $100,000 in extra reserves, you can make a loan for $1,000,000.”

    I open a bank with $100k of equity. I deposit it as a reserve at the Fed. Now I lend you $1 million which you deposit at JP Morgan. Is it your contention that I do not have to obtain $900k of funds somehow, by borrowing it, or issuing stock, or getting someone to deposit $900k? Are you saying that I can just create a $900k net asset out of thin air?

  52. Beautiful analysis; elegant writing. But what would you like us to do about it?

    “. . . We’ve got to change its preferences, which means either buying off the median influencer, or changing her identity via political struggle.”

    Either way we need to have political change to either buy off the median influencer or change her identity. So, how can we do that working against the billions now spent to shape opinion and dominate politicians?

    I think we need something like the platform described in the series beginning here: http://www.correntewire.com/a_meta_layer_for_restoring_democracy_and_open_society_part_one_conceptual_foundations

  53. Dan Kervick writes:

    K, I am assuming that the bank is already meeting its capital adequacy ratio. If it is an ordinary loan with a 100% risk weighting, then it is my understanding that if the the bank has excess capital of $80,000, it can make an additional $1 million loan. If the loan is a mortgage with a 50% risk weighting, it must have $40,000 in excess capital. If it does not have this much excess capital, it must raise more capital to make the loans. But it does not have to raise anywhere close to $900,000.

    As the banker making the loan, are not creating a net asset of $1 million out of thin air. When you make the $1 million loan, you create for yourself a liability of $1 million since you credit that amount to someone’s deposit account. In exchange you receive a pledge from the borrower for $1 million plus some amount of interest. In present value terms, your net balance is not very much affected, although if the loan is profitable for you, you should have increased your net assets. But since for the purpose of capital requirements the borrower’s pledge to you is assumed to come with a certain degree of risk, you may have to raise some additional capital, the quantity of which depends on the type of loan.

  54. Morgan Warstler writes:

    Dan, don’t be a wimp. You make MMT look bad.

  55. Dan Kervick writes:

    About what, Morgan?

  56. [...] Depression is a choice – interfluidity [...]

  57. Max writes:

    “Suppose the Fed went to negative interest rates, but refused to provide additional physical currency (though maybe they’re required to?). People could still spend through all the electronic payment systems.

    Might result in a depression for the poor, who don’t have access to those mechanisms. But everyone with debit and credit cards and paypal would be okay, no?”

    In that case FRNs would trade above face value, at a level determined by speculators (like bonds). This wouldn’t prevent FRNs from being used for transactions. It’s just that the money price of the transaction would not equal the face value of the FRNs (the goods would be “cheaper” in FRNs).

    FRNs don’t need to be money to be a medium of exchange. The moneyness (i.e. non-fluctuating value) is merely a convenience.

  58. vlade writes:

    Steve,
    Great post, and I’m really glad that you wrote it since you wrote it much better than I ever could.
    Most of the underlying problems we face are what I’d call evolutionary baggage problems. That is, we’re, to a certain extent, wired to act in certain ways – like hugely discount future, spend inordinate amount to protect what we have, assume that good times will run forever, trust people who excude confidence even when we rationally know they are liars etc. etc.
    We’re even better at rationalising all this to evil capitalists, socialists, Fed and whoever.
    We’re (as everythign living is), also great at creating positive feedback cycles (PFCs) (there’s no balance in nature, unless you count that every postive feedback has to putter out sooner or later). The important difference being, that most living things have brakes on their cycles due to clash with reality – but lots of our societal lives is disengaged with hard reality (defined as one that follows hard physical laws and hard physical boundaries/restrictions). Our “soft” reality is a much more fertile ground for PFC.

    The above sounds apocalyptic “it’s our nature, we have no choice, we’re doomed!” – but I say not so. But for us to change that, we first have to acknowledge some upleasant things – like that even as a society, we have biases, which means we end up with bad decision (and rationalise them as different means)- even thought the self-same bias might be ok for an individual, or even a much smaller society.

    So far, Kahneman and Tversky did splendid job on biases of individuals. There’s a big job for someone else to do a similar one on biases of societies (often driven by individually-rational bits) and the bad decisions it leads to. It’s a much much harder one (if nothing else, it’s hard to run experiments in the way K&T did for individuals).

  59. Ignacio writes:

    Nice post Mr. Steve. In addition to your arguments about conservative behaviour protecting bond holders and lenders in general I would like to add two other points of thinking.

    1) Policymakers may have other motivations which may be as important as defending the lenders. You mention they are risk averse and that it may me the most determinant decission driver. Another important motivation for policymakers is that they are extremely reluctant to accept that errors have been commited in the past. I mean both, errors for action or omission. For instance, you will not find eurozone politicians admitting that something was badly done in the monetary union with the exception of the idea that “Greece shouldn’t have joined”. Everything was OK ant it was the greeks, or the irish, or any other PIG who failed.

    2) Regarding the conservative elder trying to protect their own assets I think that it is correct, and my opinion is that aging in the US, Japan and Europe is what pushes the balance in favour of conservatives vs. liberals. I just hope that one day, this elder realise that they policy options, rather than inflation, is what is sinking the perspectives of the following generations.

  60. Ignacio writes:

    I don’t want to flood your blog with my messages Steve but i want to share another thinking about the conservative drift in OECD countries. I mentioned before that populations are ageing. I also believe, although I have not proof, that the political weigth of the elder is larger than that of other age segments because they tend to vote in higher proportion. That explains why policymakers do their best to defend the interests of pensioners, retired etc. while postponing measures promoting employment for everybody, but specially, for the younger joining the labour force. Fors instance, what are the top priorities for the GOP? To keep the Medicare program despite it is very expensive? To Balance government account? Reduce unemployment?

  61. “They do not pursue monetary policy with sufficient force to ensure expenditure growth even at risk of inflation. They do not purse fiscal policy with sufficient force to ensure employment even at risk of inflation. They remain forever vigilant that neither monetary ease nor fiscal profligacy engender inflation. The tepid policy experiments that are occasionally embarked upon they sabotage at the very first hint of inflation. The purchasing power of holders of nominal debt must not be put at risk. That is the overriding preference, in context of which observed behavior is rational.”

    So if more people held TIPS that would be a good thing for GDP growth and employment. Why doesn’t the government issue more TIPS, so the market is more liquid and perhaps rates don’t drop as low as negative 1%?

    For stocks, investors, it seems, should welcome inflation, greater growth, and the stocks are ownership of real assets, and so naturally hedge inflation. For TIPS, there’s little to fear from inflation, and the same for short term bonds (to a lesser extent). It’s just long term nominal bonds where there’s a problem.

    Fir TIPS it really seems like an example of the government performing one of its most advantageous functions — insurance company with unmatched dependability and economies of scale and simplicity. Who else can so easily offer widespread inflation and default insurance with unmatched dependability. It just might be a lot better if they did it a lot more.

  62. Foppe writes:

    You might like David Graeber’s Revolutions in Reverse, pp. 31-32:

    Hopelessness isn’t natural. It needs to be produced. If we really want to understand this situation, we have to begin by understanding that the last thirty years have seen the construction of a vast bureaucratic apparatus for the creation and maintenance of hopelessness, a kind of giant machine that is designed, first and foremost, to destroy any sense of possible alternative futures. At root is a veritable obsession on the part of the rulers of the world with ensuring that social movements cannot be seen to grow, to flourish, to propose alternatives; that those who challenge existing power arrangements can never, under any circumstances, be perceived to win. To do so requires creating a vast apparatus of armies, prisons, police, various forms of private security firms and police and military intelligence apparatus, propaganda engines of every conceivable variety, most of which do not attack alternatives directly so much as they create a pervasive climate of fear, jingoistic conformity, and simple despair that renders any thought of changing the world seem an idle fantasy. Maintaining this appara -tus seems even more important, to exponents of the “free market,” even than maintaining any sort of viable market economy. … Economically, this apparatus is pure dead weight; all the guns, surveillance cameras, and propaganda engines are extraordinarily expensive and really produce nothing, and as a result, it’s dragging the entire capitalist system down with it, and possibly, the earth itself.

  63. [...] interfluidity » Depression is a choice [...]

  64. K writes:

    Dan,

    I think we’ve probably gone down the rabbit hole, and maybe never comin up. But, one more try…

    We agreed (see comments 22 and 28) that to have a negative FF you have to have an equally negative reserve rate. Then at 32, you provided a long example that, apparently (though you don’t say explicitly) depends on a zero rate on reserves. That really confused me. At 40 you clarify that you are in fact assuming that you can still have a zero reserve rate. Then a whole lot of silly balance sheet discussion resulted from the general confusion. I don’t *think* we have any disagreement about bank/money accounting. (Though I do think you have a mistaken view that a bank would assume that the deposits that results from its own lending would tend to stick on its own balance sheet. Generally banks assume that new lending will be funded by a similar mix of the funding that currently makes up its liabilities, but whatever.)

    So do we agree, as we did at 22,28 that the reserve rate matches the FF rate? And for sake of discussion, can we assume that the rate on deposits also matches that rate, and that the rate on risky loans is something higher, but not necessarily positive? In which case, isn’t it also true that the Fed is free to set the policy rate wherever it wants and nothing special happens at zero?

    In case you are thinking (like some monetarists) that paying the policy rate on reserves causes monetary policy not to work for want of a hot potato… lots of CBs around the world have been doing that very effectively for a long time. The principal channel of monetary policy is to set the floor for *real* lending rates. When FF (and the reserve rate!) goes negative, lending rates will follow and there will be essentially no limit to the quantity of monetary demand stimulus that can be provided.

  65. John Berkowitz writes:

    Your article is very insightful (for me as for a non economist). I agree with you that there should´t be just the common obsession with the numbers like GDP or unemployment rates. Sometimes, our problems need deeper analysis that precedes bigger systemic changes.
    I would like to see the implication of you claim about status quo on the Canadian real estate market (the debate over More Regulation in the Mortgage Market), because I think that this approach could be very helpful in understanding, why it still haven´t changed.

  66. Morgan Warstler writes:

    Why Dan, about this:

    http://www.interfluidity.com/v2/3193.html#comment-24049

    I have been whooping you on a Guaranteed Income in multiple venues, others have noticed, and MMTers like yourself refuse to fight the subject out. It is in my Guaranteed Income vs. a Job Guarantee we learn that MMT is not economics.

    My model works PRECISELY for the reason mentions by Steve in this post. You can have half a loaf, everyone can have an income and a job.

    Why don’t you take half a loaf?

  67. [...] the key problem is the reluctance of central banks to engineer sharply negative real rates, and Steve Waldman has countered that it’s the political power of older savers that is inhibiting monetary [...]

  68. Dan Kervick writes:

    K,

    Let’s set aside the mechanisms by which the Fed engineers a market for reserves in which banks can acquire reserves at a negative cost, and let’s just accept for the sake of argument that it can do this through some combination of open market operations, discount window policies and interest on reserves policies. I think we both accept this can be done somehow. I will assume that the Fed can create a situation in which banks are in effect paid to acquire reserves. When they borrow some quantity X of reserves, they can pay off that loan in full with some quantity Y of reserves, thus netting X-Y for free.

    In fact, let’s make this this as simple as possible. Each week, the Fed sends out an email to the member banks: “Free reserves offer!” Each bank is entitled by the offer to receive some quantity of reserves free from the Fed, let’s say based on the size of the bank. The Fed credits the bank’s reserve account by the offered amount, and that’s that.

    My key point, which I made in 10, 17, and 32, is that even if the Fed pays banks to acquire reserves, or gives them free reserves, that does not give a bank one iota of incentive to issue any loans of its own at a negative rate of interest. If a bank can acquire reserves at -50% interest, or at -500% interest, or if every day the Fed reserves fairy comes and stuffs that bank’s reserve account full of massive amounts of additional reserves, for free, that does not provide the bank with any incentive to issue a loan at a negative rate of interest. There is no circumstance in which it is to the bank’s benefit to issue a liability for $X in exchange for an asset worth some fraction of $Y. No matter what the Fed does on the reserves side of the ledger to improve or degrade the bank’s balance sheet, there is never a an incentive for the bank to degrade its own balance sheet by offering a negative interest loan.

    There is no such thing as the monetary “hot potato phenomenon”, either in the narrow bank market for reserves or in the broader market for money. Since there is no cost in the modern world to holding a vast monetary “inventory” of any size, there is no reason at all for a holder of money to pay someone to take some of that inventory off their hands – which is what a negative interest rate loan is. I have argued against the hot potato thesis several times at Nick Rowe’s blog, and so this is not the place to repeat that argument. But here is my most recent version.

  69. [...] Waldman knocks it out of the park again with his new piece called “Depression is a Choice“. He’s like Babe Ruth without the [...]

  70. [...] Source [...]

  71. Philo writes:

    “They do not pursue monetary policy with sufficient force to ensure expenditure growth even at risk of inflation. They do not pursue fiscal policy with sufficient force to ensure employment even at risk of inflation.” Literally, what these sentences say looks quite unobjectionable—i.e., true. But they suggest something dubious: that monetary policy has a direct effect on the level of spending, but at most only an indirect effect on employment; while fiscal policy has a direct effect on employment, but at most only an indirect effect on spending. This looks objectionable—i.e., false. If you did mean this, let me ask why you believe it.

  72. Dan Kervick writes:

    There is no circumstance in which it is to the bank’s benefit to issue a liability for $X in exchange for an asset worth some fraction of $Y.

    Sorry about the confusion K. Read $X in both locations in that sentence.

  73. K writes:

    Dan,

    “In fact, let’s make this this as simple as possible…”

    Why are you inventing a whole new mechanism? That’s just going to obscure and confuse the discussion. Instead, I’ll make it easy for you to refute my argument. Just assume NOTHING CHANGES, except:

    1) eliminate cash
    2) pay interest on reserves at the Fed Funds rate
    3) set the FF negative, lets say -5%

    I’ll make two clear contentions for you:

    1) banks will lend to each other at the FF rate. The lenders will pay the borrowers 5% to *take* their money.
    2) banks will lend to their good clients at a rate just above -5%, but well below zero.

    Show me why they wont. In your answer please make sure to address the following issues:

    1) borrowing at -5% and lending at -4% is profitable
    2) reserves pay *negative* interest. There is *no* instrument that pays 0% interest.
    3) A bank can’t borrow at -5% and *not* put the money somewhere. Where do they put it?

  74. Morgan Warstler writes:

    wimp

  75. Greg writes:

    So what IS economics Mr Warstler? Youve already described the status quo as fascism, where the real asset holders are just puppeting the politicians and getting their pound of flesh from everyone else. That makes the Sumners et al simple front men for the few who absolutely will not have things being bought and sold on terms other than their very own, modern day slave drivers.

    It seems there IS no economics. It all is a front for what is really happening There is no forces of supply and demand it s jut a few mega trillionaires fighting each other for “The most zero’s after the nine” in the history of the planet.

  76. Dan Kervick writes:

    Morgan, I’ve tried to avoid being blunt, but let me now be blunt. I don’t know if you are an a-hole in real life or just play an a-hole on the internet to entertain yourself. But from long fatigue with your act, I now skim past just about everything you write. Maybe that’s my loss, but there you have it. If you think you have a great proposal then send it to the New York Times or someone else with more patience for your routine. I know you are some kind of great salesman and wealth creator, so I’m sure you will have no trouble getting more important people than me to pay attention to you.

  77. Greg writes:

    Pretty sure he’s an A hole in real life. No non A hole would get great joy out of playing an A hole on the internet.

  78. Morgan Warstler writes:

    Dan, you are a wimp. Skip reading that.

    I have a basic proposal it delivers EXACTLY what your wing of MMT says it wants – a JOB for everyone.

    A JOB for everyone. Virtually overnight.

    You have a legitimate counter-plan in front of you, from someone who takes the time to read what you write, that is far more politically likely AND actually scales, no shovel ready job problem.

    My point is, you can call me a grade A asshole all you want, but there is a REASON you don’t like my proposal, and you really don’t want to get into it, it exposes your deeper goals, which are not economics.

    And that’s how good I am Dan, I’m here to trap you into saying that Econ for you is about Dan having more power.

    Greg, woot! for you

    “Youve already described the status quo as fascism, where the real asset holders are just puppeting the politicians and getting their pound of flesh from everyone else.”

    Dude, econ is supply and demand. Modern econ is that digital is not scarce, and the atomic is scarce. And slowly more of life is digital.

    I’m NOT insisting on what you call fascism, I’m saying if you don’t RECOGNIZE the realities that Steve himself talks about about all the way down to the bottom, you can’t REALLY craft econ policy.

    Instead you end up making like Dan, who claims to want to do econ policy, but NOT REALLY, he wants to topple the apple cart.

    And you can WANT to topple the apple cart, but that’s not ECONOMICS Greg.

    Economics tries to be about what really IS, it wants so very desperately to be a science. And if you start out with an agenda first, and then scaffold some econ chatter around it, that’s not Econ.

    What I want is policy proposals that CAN BE ADOPTED in the real world.

  79. Vinz Klortho writes:

    To quote Morgan: “Economics tries to be about what really IS, it wants so very desperately to be a science”

    Any endeavor that has as one of it’s maxims that “individuals are rational” has nothing
    to do with “what is”.

    Vinz

  80. Dan Kervick writes:

    K, it seems to me that perhaps we cannot understand one another because you are using a model in which banks lend their reserves, so you are assuming that when it makes the loan, its money-losing reserve account is drained by the amount of the loan. But I don’t believe that is what happens. Let’s go through the loan scenario step-by-step.

    Suppose Bank A currently has $10,000,000 in deposits and $1,000,000 in reserves. Suppose IOR is -5%, so that the Fed is basically taxing the bank 5% on its reserves. If Bank A holds its reserves at exactly the current level throughout the year, it loses $50,000 to the Fed. It can also borrow additional reserves in the Fed Funds market at -5%.

    Suppose Bank A now makes a one-year loan of $1,000,000 at -4%. It creates a deposit balance for the loan customer of $1,000,000. The customer in return is pledged to pay the bank $960,000 by the end of the year to discharge the debt. Since the deposit it has just created is a liability, the bank must increase its reserves by $100,000 to meet its reserve requirement. It does so buy borrowing the reserves in the interbank market at -5% from Bank B. Now there are two main scenarios to consider:

    1. Let’s assume first that all the checks the loan customer writes on the deposit are to other customers of the same bank, who just save the amounts they are paid. in that case, the deposit balance is just divided up and moved around from account to account throughout the year, but the bank’s total deposits don’t change. So the bank has to hold the additional reserve balance of $100,000 throughout the year. Bank A borrowed that balance from Bank B, and Bank B pays Bank A $5000 as a result in line with the FF rate of -5%. But the Fed charges Bank A $5000 for holding the reserves for a year in line with the -5% rate of interest on reserves. These amounts cancel. At the end the year, the loan customer delivers to Bank A the required check for $960,000 drawn on an account from bank C. As a result, a Fedwire transfer of $960,000 is made from the reserve account of Bank C to the reserve account of Bank A.

    What is the net result of these transaction on the balance sheet of Bank A?

    a. +$5000 from Bank B
    b. -$5000 to the Fed
    c. -$1,000,000 in additional deposit liabilities
    d. +960,000 from Bank C
    TOTAL: -$40,000

    So making the loan actually cost the bank money. It did not benefit from making that loan.

    2. Now let’s assume now that as soon as Bank A makes the loan, the loan customer writes a check for the whole $1,000,000 to buy something, and that check is deposited in Bank D. The $1,000,000 deposit balance that the loan customer was given at Bank A is cleared, and a $1 million Fedwire transfer is thus made from the reserve account of Bank A to the reserve account of Bank D. At the end of the year, just as before, the loan customer delivers to Bank A the required check for $960,000 drawn on an account from bank C. As a result, a Fedwire transfer of $960,000 is made from the reserve account of Bank C to the reserve account of Bank A. Because the $1,000,000 deposit balance that was created for the loan customer is cleared right away, Bank A never has to borrow any additional reserves.
    Now what is the net result of these transaction for for Bank A?

    a. -$1,000,000 to Bank D
    b. +960,000 from Bank C
    TOTAL: -$40,000

    So again making the loan actually cost the bank money. It did not benefit from making that loan.

    Now let’s imagine something much more radical. The Fed gets rid of the interbank market and IOR. All that is left is the discount window. And the Fed sets the discount rate at -5%, with some annual borrowing maximum. The Fed is paying banks to borrow money! And now let’s assume the other starting conditions for Bank A are the same: $10,000,000 deposit balance; $1,000,000 initial reserves; and the bank makes a $1,000,000 loan at -4% interest, requiring it to boost it’s reserves by $100,000. (Assume scenario 1 from the above.) It borrows these reserves from the Fed at -5% interest. What is the result?

    a. +$5000 from the Fed
    b. -$1,000,000 in additional deposit liabilities
    c. +960,000 from Bank C
    TOTAL: -$35,000

    It hasn’t lost as much as before, but it has still lost.

    Now, assuming the Fed borrowing limit is very high, the bank can still make money if it borrows lots of excess reserves. So suppose it borrows $1,000,000 in reserves at -5%, not $100,000. Then here is the result:

    a. +$50,000 from the Fed
    b. -$1,000,000 in additional deposit liabilities
    c. +960,000 from Bank C
    TOTAL: +$10,000

    The bank made money! But suppose the Bank had simply decided to borrow the $1,000,000 in reserves from the Fed at -5% and had not made the loan. Then the result is this:

    a. +$50,000 from the Fed
    TOTAL: +$50,000

    Permute things how you will, there is no scenario in which making the loan at negative interest makes money for the bank. For any combination of actions in which it makes that loan and makes money, there is some alternative combination in which it refrains from making the loan and makes even more money. And for any combination of actions in which it makes that loan and loses money, there is some alternative combination in which it refrains from making the loan and loses less money.

  81. K writes:

    Dan,

    “you are using a model in which banks lend their reserves”

    Nope. Pretty sure that’s not it.

    1. You forgot

    e. +$50,000 from -5% interest on -$1,000,000 deposit balance

    TOTAL: +$10,000

    2. When Bank A made the $1,000,000 Fedwire payment to Bank D, Bank A had to go borrow that money in the Fed Funds market (Effectively that loan will come from Bank D). Bank A earns $50,000 for taking that loan for 1 year at -5%.

    TOTAL: +$10,000

    The quantity of required reserves have nothing to do with it. Could be zero, could be 10%, doesn’t matter. As you say, you can just borrow the reserves in FF market, and it’s a wash.

  82. K writes:

    Dan,

    In your third example you forgot

    d. Deposits said reserves at Fed: +$5,000
    e. interest on deposits: +$50,000

    TOTAL: $10,000

    In the fourth example… What did the bank do with the borrowed money until it repaid the loan?

  83. K writes:

    Oops:

    Correction and clarification

    “d. Deposits said reserves at Fed. Interest: *-$5,000*
    e. Interest on deposit liabilities: +$50,000

    TOTAL: $10,000″

  84. [...] U.S., Europe and Japan have share one big problem: protecting incumbent creditors – Interfluidity How Abbott Labs keeps the U.S. from saving $700 million year – The Incidental [...]

  85. Dan Kervick writes:

    1. You forgot

    e. +$50,000 from -5% interest on -$1,000,000 deposit balance

    K, how does the bank earn interest on its own customers’ deposit balances? Those balances are bank liabilities.

    2. When Bank A made the $1,000,000 Fedwire payment to Bank D, Bank A had to go borrow that money in the Fed Funds market (Effectively that loan will come from Bank D). Bank A earns $50,000 for taking that loan for 1 year at -5%.

    TOTAL: +$10,000

    First of all, it doesn’t have to borrow any additional reserves to make that payment, since we assumed its starting reserves were already $1 million. But more importantly, per your assumption of -5% IOR, for any additional reserves Bank A borrows, it makes money from the lending bank but owes money to the Fed.

  86. K writes:

    Dan,

    “how does the bank earn interest on its own customers’ deposit balances? Those balances are bank liabilities.”

    That’s the meaning of negative rates. The lender pays the borrower.

    “First of all, it doesn’t have to borrow any additional reserves to make that payment, since we assumed its starting reserves were already $1 million.”

    OK. Then there are $1 million less reserves which saves the bank $50,000 in interest.

    “for any additional reserves Bank A borrows, it makes money from the lending bank but owes money to the Fed.”

    Except that they *didn’t* deposit that money at the Fed, right? They Fedwired it to bank D.

  87. Dan Kervick writes:

    That’s the meaning of negative rates. The lender pays the borrower.

    That’s begging the question in the matter of the customer’s deposit balance, K. I started by saying that Yglesias hadn’t thought through the mechanisms of getting to negative rates, and you said that the target FF rate would do the trick. I said it wouldn’t do the trick and reasoned through the mechanisms, and suggested that much more comprehensive regulations would have to be imposed. But now you’re just saying in challenging that reasoning, “Assume all interest rates are already negative.”

    It still seems to me that you are employing a model of banks lending their reserves. You are assuming that if the Fed creates a system in which a bank is taxed for holding reserves – because of negative IOR – then a bank will want lend out those reserves at any a rate above that negative IOR rate so that it loses less via the loan than it does by holding the reserves in its account with the Fed. But that’s not how it works. Banks don’t lend their reserves. If a bank is not already carrying excess reserves, then for every additional dollar in deposit balances it creates via loans, it has to acquire additional reserves. That’s not just in a US-style system with a mandatory reserve ratio. The greater the volume of deposit balances a bank’s customers possess, the greater the volume of reserves a bank needs to hold in order to accommodate the higher volume of payments it will need to make as its customers conduct transactions with their deposit balances. The bank’s reserve account at the central bank is the account through which those payments are settled and cleared.

    Remember, we are assuming no physical cash. So the only way businesses and households can possess money in this scenario is in the form of a bank deposit balance – points on a bank scorecard. If the banks are expanding their balance sheets by making loans, then their reserves balances have to be growing as well.

    But the more I think through this, the clearer it becomes how utterly absurd Yglesias’s underlying hypothesis is. He actually seems to think that the whole reason we have recessions is because there happens to be one specific kind of non-maturing, zero-yield, financial asset – namely, physical currency – that people have the option to acquire in exchange and hold. But how does the existence of such an asset affect the underlying logic of financialization, debt and over-leveraging? Financial fragility and the pyramid of debt results from people attempting to borrow at a lower rate of interest from the rate at which they lend. If people were in the practice of trying to make loans at -3% financed by borrowing it -5%, then all the same problems of potential insolvency would still exist. Miscalculation of risk, irrational exuberance, and sheer fraud will still exist. Ever more intricate and fragile systems of contracted cash flows still exist. What is to prevent a Bernie Madoff in such a system from from promising vast and unheard of astronomicalreturns of 2% or 3%, and making that system work for a while through Ponzi finance by convincing some suckers that he is really, really clever? What is to prevent a top-heavy financial system of derivatives created to bet on the risks inherent in the first layer of the system?

    And if the debt contracts are still written in nominal terms, and if following a credit event society still insisted on the creditors either getting paid what they are owed or repossessing loan collateral, then we would have all the same problems of long, stagnating deleveraging and unemployment that we have now. And if the central bank solution were to somehow push rates into deeper negative territory so that everybody had a strong incentive to spend money as soon as they acquire it, then we would have massive dissaving and it would be even more impossible for households to repair their balance sheets, as the gap between the interest one could could earn through investment and the fixed interest charges on one’s debts would grow larger.

    Finally, it strikes me that the whole system would be massively deflationary. The only reason the private sector’s net supply of monetary assets is able to grow is because the monetary authority continually injects additional money in to the system. But for this negative interest rates scheme to work, the monetary authority would have to continually extract monetary assets from the economy. This would result in wage and price deflation, and people’s fixed nominal debts burdens would quickly result in mass insolvency.

  88. Morgan Warstler writes:

    Dance around it all you want Dan, you are hiding from me. Don’t wimp out Dan, just step up and prove why my idea isn’t more likely than yours.

  89. Travis writes:

    Again a solid outing with a bent reflection. The present policies are not the outcome of an ageing population that is risk adverse. They were dreamt up in the 50s 60s and 70s by very youthful and vigorous men and some women. What you have described inter alia is the mechanism by which the risk takers (not) fleece the risk adverse. And every fleecing leaves the risk adverse begging for a near risk free product that offers them some exposure to to the up side and none of the down side to make up for the last loss. Hence why the GFC came originated in AAA packaging.

    Please, please, stop using your big brain to tell determinist stories, demographic or otherwise. By all means psychologise the Muppets but be sure to tell a story about their puppeteers. I get that old people are scarred about violent crime but what is more interesting is why they think it has been increasing not decreasing over the last thirty years.

  90. K writes:

    Dan,

    “But now you’re just saying in challenging that reasoning, “Assume all interest rates are already negative.””

    No, I’m not.

    Since the end of Reg Q, deposit rates have been tracking Fed Funds. Since we hit the zero bound some have even gone negative. The mechanism for deposit rates to follow Fed Funds below zero are already in place. *That* is what I said. I never said rates were already negative.

    ” You are assuming that if the Fed creates a system in which a bank is taxed for holding reserves – because of negative IOR – then a bank will want lend out those reserves at any a rate above that negative IOR rate so that it loses less via the loan than it does by holding the reserves in its account with the Fed”

    No, I’m not.

    Banks don’t lend reserves, they create money by lending. Banks are capital constrained. Reserves are completely irrelevant to this conversation. There is *nothing* in anything I said above that depends in any way whatsover on the role of reserves, mandatory or otherwise. Unless you can point to a single incorrect statement that I’ve made, don’t imply that I don’t understand how the system works.

    “And if the central bank solution were to somehow push rates into deeper negative territory so that everybody had a strong incentive to spend money as soon as they acquire it, then we would have massive dissaving and it would be even more impossible for households to repair their balance sheets”

    Yup. See the last paragraph of my very first comment (7) above.

    I see you want to move the conversation along. I’m all for that; in fact it’s what I’ve been wanting to do since yesterday morning at the end comment 7. The only obstacle was your dogged determination that negative rate lending is impossible. So I spent two days, carefully, and as politely as possible, picking apart every incomplete example, wrong statement, incorrect calculation and red herring that you could throw at me. Parrying your serially mutating objections has been an experience akin to a Whack-A-Mole super marathon. And now, after apparently changing your mind (silently) a half dozen times, you try to insinuate that it is I who has some basic misunderstanding of banking. It’s annoying. The least you can do is to acknowledge the incorrect statements you’ve made (and *perhaps* the fact that you may still have a bit of thinking to do about how banking works), rather than just try to inconspicuously move on and pretend you didn’t drag me through this for two days straight.

    You still haven’t explained how the bank made $50,000 in your last example in comment 80.

  91. [...] Volcker moment.  He has had many opportunities and whether because of groupthink at the Fed, political power of savers, or a failure by him to read Scott Sumner’s blog, Bernanke cannot seem to find [...]

  92. [...] up which to a doughnut salesman looks like a sublime pastry of a post.  SRW tells his readers that Depression is a choice.  He is on the money with title.  I guess I have two problems with the post the first is purely [...]

  93. [...] interfluidity » Depression is a choice [...]

  94. Dan Kervick writes:

    You still haven’t explained how the bank made $50,000 in your last example in comment 80.

    I thought that was the most straightforward one, K. The discussion of the third and fourth examples started with the sentence “Now let’s imagine something much more radical.” The more radical scenario I then stipulated was one in which there is no longer any Fed Funds market and no longer any interest on reserves, and in which banks acquire reserves by borrowing at negative interest from the discount window. So then, in the last example, the way the bank makes $50,000 is easy: it borrows $1 million form the Fed at -5% interest. It then pays back $950,000, fully discharging the debt. That’s all it does. It has netted $50,000 in the deal.

    The reason I considered that hypothetical scenario is because I have been trying since the beginning of this discussion to zero in on some way to imagine a set of policies in which banks can “fund at a negative rate”, which I thought was where you began. A situation in which there is both a negative Fed Funds rate and an equivalent negative interest rate on held reserves is not such a mechanism, since if a bank acquires reserves by interbank lending and holds them to satisfy the reserve needs created by the loan it makes, the bank in effect has to pay back to the Fed the money it makes from borrowing the reserves.

    So what I have been trying to argue is that even if a bank can fund at a negative rate, it is not then in the bank’s interest to make a commercial loan of its own at a negative rate. The simplest way of understanding that, it seems to me, is to imagine that the way the bank funds is by borrowing reserves from the discount window at a negative rate of interest. When you boil the situation down to that, which I think is actually the most favorable scenario for your case, it is pretty clear that the bank does better by borrowing the reserves and not making the loan (4th case) than it does by borrowing the reserves and making the loan (3rd case).

    The reason I have shifted the examples is not because I have been changing my mind, but because when we have reached a stumbling block where we didn’t agree, but which I thought wasn’t essential to that main point I was trying to make, I tried another approach to zeroing in on the main point.

    Reserves do matter, because making additional loans typically requires that banks expand their reserve holdings. So in determining whether the decision to make the loan is profitable for the bank, we have to calculate all of the effects of making the loan. I have argued that even if we imagine situations in which the bank benefits or breaks even from acquiring those reserves, that part of the operation does not offset the loss of making a negative interest loan.

    I always learn more about these kinds of operations every time I sit down to think hard about them, K. But on the question of whose picture of banking operations is evolving more as we go along, could I remind you that you were extremely insistent earlier in the thread that to make the $1,000,000 in new loans with $100,000 in excess reserves, the bank needed to find $900,000 more in “funds”, and expressed shock that I thought the bank could create those deposits for its loan customers out of thin air. But then after we talked about capital adequacy, you seemed to drop that line of thought, and now accept that the banks “creates money by lending.”

    I’m exhausted and have a cold, so you can go ahead and have the last word.

  95. K writes:

    Dan,

    I don’t want the last word. I just think it’s courteous to acknowledge your mistakes so the discussion can move on. But I’m sorry I was rude.

    I *still* say that banks have to find the $900k because the newly created deposits are dispersed throughout the banking system. People don’t take out high interest loans so they can keep the new money as low interest deposits. Generally the deposit is alienated from the lending institution the moment the loan is made. So the bank then borrows back the funds. The system as a whole creates its own deposit funding, but any given bank does not. That is my working scenario, and it hasn’t changed a bit. I was shocked because I thought you were saying banks didn’t have to have liabilities (loans or deposits) to fund their assets. I’m still confused by the apparent lack of an asset in your fourth example.

    In your fourth scenario you *still* haven’t told me what the bank does with the money it borrowed. We’ve created a debt, but there’s no offsetting asset! If the bank doesn’t hold the money in the form of some instrument over the term of the loan, how does it pay back the loan? I really don’t get it.

    Sorry you are sick. Hope you wake up feeling better.

  96. Morgan Warstler writes:

    Dan, don’t play sick bed. Don’t be a wimp. I’ll meet you in next thread. #bested

  97. [...] Randy Waldman has a characteristically excellent post where he argues that the monetary policy we observe in Japan, the Eurozone and to a lesser extent [...]

  98. Narg writes:

    Very clear-headed article. I read a lot and this is the best summary I’ve ever seen of what happened over the last five years.

  99. The situation is a choice because and only becasuse of the following reasons:

    1) the Keynesian Spenders, Chicago Monetarists, Classical Liberal Industrial Policy Advocates and Austrian Human Capital Advocates will not, together, put suite of solutions together that will provide each ‘economic political party’ and its constituency with compensation for the involuntary transfers that will occur, and the negative externalities that will be brought about, if we borrow and spend.

    I blame this entirely on Krugman, Yglasias, yourself and the entire mainstream movement whose ‘party’ is in power, and who, like all parties in power, seek to push their agenda independently of compromise rather than that of the collective through compromise. The people in government do not have a sufficient grasp of the different schools to think of them as the adjuncts to political parties that they are.

    It would be entirely possible to ‘spend’ in exchange for wiping out the DOE, HUD and public education tenure. That would be a fair exchange. It would be entirely possible to ‘spend’ in exchange for a new immigration policy. That would be a fair exchange. But all efforts at exchange have failed. Polarization continues. And you simply seek economic dictatorship, so that you can remove the means by which the population can rebel against the state.

    2) I agree with the MMT crowd, and Yglasias, that the elimination of cash money will allow forcible redistribution by way of monetary policy alone, which will allow the Statist/Keynesian party to overwhelm the Monetarist, Industrial and Human Capital parties, and each of their supporters in the economic legislature.

    3) The other Political/Economic party coalitions object, because they object to further empowering the state/keynesian party. This is the opportunity that the other coalitions are using to punish the state for over reaching. The conservative strategy is to starve the beast and bankrupt it before it can bankrupt them.

    4) It appears to most of us, who focus on productivity instead of consumption, that both increases in spending, and a cashless society simply removes the constraints on destruction of productivity and the creation of catastrophic bubbles that will not be able to be ‘fixed’ by corrections, but instead, by revolution, economic or military conquest.

    So yes, People clearly prefer this state of affairs to that in which the state is further empowered to expose them to risk

    And in that sense, it is a rational choice, a fair trade, and it is currently being purchased at a discount.

    There is no difference between economics and politics.

    Economics is a subset of politics.

  100. STATUS and NORMS are higher value assets than money and credit.
    Money and credit are just tools to obtain status and establish norms.
    A diverse population cannot cease conflict over status and norms.

    Failing to understand this distinction is the error of financializing a polity.

    It is a professional cognitive bias.

  101. Peter K. writes:

    I liked this blogpost. Today’s depression is similar to the Great Depression in that monetary policy was a failure and too tight. The jaunty FDR gave the bird to conventional wisdom of the day and went off the gold standard. (The catholic FDR also tried a bunch of fiscal and job policies some which worked some of which didn’t). And then finally there was rearmament in preparation for WWII.

    The early 80s downturn was solved with monetary policy as was the early 90s downturn and the post tech bubble of the late 90s. Today we’re facing the zero bound and the cult of low inflation which is like the gold standard of the 1930s.

    Debating Dan K. is a fool’s errand. Some American voters don’t understand that having full employment and an economy growing at capacity (without runaway inflation) is in the majority’s interest. It’s why the Fed has a dual mandate. Unlike Europe and Japan, the Fed is America’s safety net. Dan K. and those like him don’t want the Fed to even try to do more.

  102. @PeterK

    It’s not a cult. It’s a program. It’s a strategy. That conservatives speak in allegorical moral language is one thing. That the underlying strategy is entirely rational is something else. The strategy is, was and will remain consistent: force the bankruptcy of the state, as a means of establishing terms for negotiation on a compromise settlement. That has been the strategy since the late seventies, when the right gave up hope that ‘wisdom’ would prevail. The right offers these terms consistently. If they are unacceptable, then this process of polarizing conflict will simply proceed until one side wins totalitarian control of the state or the other. And that will be followed by an even greater conflict. This is the stuff of which civil wars are made.

    From the standpoint of geopolitics and political philosophy, the arguments of economists are amateurish. Human beings behave consistently over time. They are more pacifist in wealth, but that is all that changes. The human economy is status signals. Money serves them.

    Curt

  103. [...] Is an Economic Depression a Choice? Japan, it seems [...]

  104. studentee writes:

    Good post Steve, and good back and forth between Dan and K.

    “Debating Dan K. is a fool’s errand.”

    Dan Kervick usually brings something originally and thoughtful to every discussion I’ve seen him take part in. You on the other hand, just seem to follow him around, dropping awful, consensus-mongering prose.

    “The early 80s downturn was solved with monetary policy as was the early 90s downturn and the post tech bubble of the late 90s.”

    Both more likely solved with, respectively, the Reagan deficits and the Bush tax cuts.

    “Unlike Europe and Japan, the Fed is America’s safety net.”

    Stunning.

  105. Morgan Warstler writes:

    Dan, don’t make lefties look weak. Explain to the taxpayers who fund the govt. why as a safety net, the govt. shouldn’t give the unemployed their income on Paypal and auction their labor on Ebay.

    C’mon, Dan you love to preach Job Guarantee.

    In all things Dan, when you open the door, your opponents, your betters from the other side, will march right through. You say Jobs Guarantee, I say Guaranteed Income / Auction the Unemployed – and you duck, duck, duck.

    Don’t be a wimp Dan, it makes MMT look weak.

  106. Foppe writes:

    I admit that I don’t always read the comment threads here, so I don’t know if ‘contributions’ like Morgan’s are a frequent sight or not, but damn..

  107. paul writes:

    @Curt Doolittle

    “It appears to most of us, who focus on productivity instead of consumption…”

    What utter bullshit…you have no idea what makes an economy move.

    What is the point of production without consumption? We might as well be back living off the land. The stuff you produce would have no value if the rest of us didn’t have the means to obtain it. Do you sell your stuff to China’s workers?

    “…destruction of productivity…” – more nonsense. So-called producers extract wealth out of others productivity. Why should I produce for them? It’s a one-way mathematical dead-end if some entity isn’t monetizing the system to keep it moving, all the while the “producers” are extracting wealth at other’s expense.

    Increasing productivity is to increase unemployment. When one person can produce for the other 7 billion what will we need the 7 billion for? The end of humanity will come way before then of course. Capitalism will have no meaning in a fully mechanized society. Machines are ultimately capable of performing any task that humans can.

    Bubbles are caused by greed and ignorance of the masses as to how they are being used by the parasites masquerading as “producers”. Wealth accumulation is what creates the need for massive spending. Without spending as it’s been over the past 30 years most of us would be starving. As it is nearly half the population is at or barely above poverty.

    I’m sick and tired of a small group of moron “producers” that were born on third base telling me that they hit a home run and the rest of us need to get off our asses.

    “…STATUS and NORMS are higher value assets than money and credit.…”

    Yeah for 0.01% of the population. Everyone else just wants to be able to have a roof over their head, be able to eat and raise their families, all the while trying to enjoy life, whatever that means.

    I don’t give a damn what that 0.01% wants.

    Your simpleton world-view ignores fundamental laws of the universe, human nature included. Norms are not universal and never will be. Your norms aren’t interesting to me and you will achieve them over my dead body. I choose my own norms.

    btw credit is not an asset – it is identically equal to zero on the day you receive it. Less-than-zero as the interest accrues.
    The liability lives on long after the joy of the asset is gone. I see a place for credit re business investment. As an enabler of consumption it is like an addictive drug.

    Money is nothing more than numbers denominated in the unit of account. It’s a scorekeeping system that is necessary for an economy to have enough motion to sustain a population.

    Without something like money your dream world economy doesn’t move – it sits there like the sand at the bottom of an hourglass, crickets chirping…

    Go crawl back up Ron Paul’s ass.

  108. beowulf writes:

    Morgan, you should to patent your ebay hiring hall and start marketing it to states. You can thank me ($$$) later.

    “The Obama administration is looking for states that will experiment with unemployment insurance programs by letting people test a job while still receiving benefits. The plan is a key feature of a payroll tax cut package that President Barack Obama negotiated with congressional Republicans in February.

    The Labor Department will open the application process Thursday for 10 model projects across the country. Any state can apply for the “Bridge to Work” program. The plan is modeled after a Georgia program called “Georgia Works.” Under the plan, workers who have lost jobs can be placed in other temporary jobs as trainees for short periods to retain their skills or gain new ones while receiving jobless assistance.”
    http://news.yahoo.com/states-asked-apply-unemployment-test-plan-125932074.html

  109. @paul

    RE: “Go crawl back up Ron Paul’s ass.”
    thank you for the articulate recommendation, and the emotional invective. Oh. Wait. The bag is empty….

    RE: “What utter bullshit…you have no idea what makes an economy move.”
    Again, thank you for the articulate criticism… Hmmm… Wait. That bag is empty too….
    Hmm… lets see. Inter-temporal misallocation of capital because of interference in production cycles not the least of which is the maturity of the production cycle of credit.
    I think I understand that over any period of time, an economy is not a closed system, and that human learning and therefore human lives depend on not misallocating capital.

    RE: Do you sell your stuff to China’s workers?
    Um… productivity is a measure of allocation of capital, particularly human capital. We must innovate to compensate, so to speak.

    RE: “So-called producers extract wealth out of others productivity. ”
    Wait, does this little paper sack here have the remaining cookie crumbs of the labor theory of value? I think so. Yes. that’s certainly what it looks like. Spoiled by the smell of it.

    RE: Increasing productivity is to increase unemployment.
    Wait, do I hear the belief that skills and knowledge and ability are neutral? That allocating capital to consumption of goods that require uncompetitive skills doesn’t distract people from learning competitive skills? Hmmm… I think that’s what I hear. Yes. THat’s certainly what I hear. No wonder we can’t get all economic political parties to address the full socioeconomic production cycle, from the immediate spending to short term monetary policy to linger term industrial policy, to generational human capital policy. I guess we have to move ONE lever at a time, we can’t move all four levers.

    RE: “Bubbles are caused by greed and ignorance of the masses as to how they are being used by the parasites masquerading as “producers”.”
    I thought your name was paul not Karl.

    RE: “I’m sick and tired of a small group of moron “producers” that were born on third base”
    Well, I guess those of us who fought our way up the ladder are a little sick of over people’s hands in our pockets. No on had a worse starting place than I did.

    RE: “…STATUS and NORMS are higher value assets than money and credit.…” Yeah for 0.01% of the population.
    You mean, that you want status too. ‘Cause shelter and food are pretty cheap.

    RE: :I don’t give a damn what that 0.01% wants.”
    Well, it’s actually about the 50%. which is what the polls show.

    RE: “Your simpleton world-view ignores fundamental laws of the universe, human nature included. Norms are not universal and never will be. Your norms aren’t interesting to me and you will achieve them over my dead body. I choose my own norms.”
    Um. That’s not a logical statement. Kinda funny. I should save that one. It’s not a norm if you can choose it. that’s why we call them norms. That’s right up there with “I have my own truth”.

    I never tolerate ad hominem’s. It’s bad for society.
    OK That was a fun bit of silliness, but a complete wast of my time. So back to work now. Ta’.

    :)

  110. paul writes:

    @Curt

    Thanks for reading… :-)

    Needless to say I don’t have much respect for your worldview. You caught me on a day when I don’t have a lot of patience for your brand of drivel.

    “I think I understand that over any period of time, an economy is not a closed system” – Curt

    Clearly you have little understanding of closed systems. Time has nothing to do with the closed property. The domestic private economy is a closed system and can be modeled as such with a high degree of confidence. It follows all of the laws of closed systems when viewed from the proper perspective. One can define a closed-system boundary as one wishes and work within that definition. Over all of history or for 5 minutes..

    “No one had a worse starting place than I did”. – Curt

    Were you born in a ghetto?

    “We must innovate to compensate” – Curt

    “Endeavor to persevere” – Lone Watie, The Outlaw Josie Wales

    Lone Watie makes more sense than you do.

  111. Morgan Warstler writes:

    Foppe,

    Just explain why I’m wrong. Dan can’t say out loud why he doesn’t like my plan.

    If you start explaining it, there will be a moment where you go silent, and that’s why Dan doesn’t answer.

  112. Foppe writes:

    Morgan: I am not even going to pretend that your proposal makes sense to me. It is so laden with assumptions (most importantly, your suggestion that everyone who doesn’t do wage labor is socially useless) that I wouldn’t have a clue where to start unraveling it.
    Having said that, even if your suggestion did pass even the most cursory of smell tests, it still wouldn’t excuse your trolling.

  113. Dan Kervick writes:

    Morgan, where have you written up your plan in detail. I don’t really know what it is.

  114. K writes:

    Dan,

    Since you are only talking to Morgan now, let me answer my own question. The implied asset in your example, of course, is *reserves* which cost $50,000 to hold. This, not coincidentally, leaves the bank with a profit of zero. If they had lent the money to a client at -4%, the profit would have been $10,000. The theme here is $10,000. No matter how the bank funds the loan (deposits, Fed Funds, discount window) the profit of lending at -4% is $10,000. Why? Because the bank liabilities all pay -5%.

    Though in your four examples you got four different wrong answers, there is a common theme to your errors. In every case you failed to ensure that there were the same quantity of assets and liabilities. It’s the same thing that you did in comment 53, when you said that a bank that gives a loan to a customer who transfers the money to JP Morgan doesn’t have to raise funds to cover the loan, and instead just has to ensure that it has adequate capital. This, along with your persistence in claiming that your fourth example makes sense, is what made me think, and continues to cause me to think, that you don’t believe that banks need to fund their assets or back their liabilities.

  115. JW Mason writes:

    Brilliant post, exactly right.

    The ZLB is a huge distraction — as you say, there is no more reason to think that monetary policy, let alone demand management more broadly, is ineffective now than at any other time.

    That said, it is probably true that while the ineffectiveness of policy today is exaggerated, the effectiveness of policy in the past was probably also exaggerated. A generation of economists and policymakers have been sold on the idea that the central bank can reliably and costlessly keep demand at the optimal level (which itself can be unambiguously specified.) This was never true, and it’s become steadily less true as the regulatory framework within which postwar monetary policy operated (reserve requirements, etc.) has been eroded.

    So while you are certainly write in the larger sense that depression is a choice, it is also the case that this choice is easier to make because the traditional tools of depression-prevention don’t work as well as advertised — again, not because of the ZLB, but because (a) it is hard for a central bank to control the pace of credit creation in a deregulated financial system and (b) the availability of credit is not the sole or decisive influence on private expenditure decisions. Developing better tools is certainly possible, but the need to do so makes the hurdle for an anti-depression politics higher.

  116. Morgan Warstler writes:

    Guaranteed Income / Auction the Unemployed

    Using a clone of Paypal and Ebay platforms, the US govt. should establish a Guaranteed Wage of $240 per week. Anyone who wants to work registers, receives a Debit Card,and each Friday has their GI deposited.

    All recipients have their labor weeks auctioned online. Bidding begins at $40 per week ($1 per hour). Bid increases by .50 cents per hour ($20 increments).

    Recipients keep 50% of the top bid, if they take it. If they opt for a lesser bid outside certain boundaries there are penalties (fraud measure).

    Recipients cannot be made to work outside a radius of a couple miles.

    Bidders must deposit money into system before they bid. They must accurately describe the job. Feedback will be given both ways. If you are familiar with Ebay, you understand what this accomplishes.

    There are no taxes paid, there are basic workplace protection requirments. Umbrella insurance is sold on site for folks bringing labor into their home.

    Expect 30M to register so approx $345B is our cost assuming 30M are auctioned at $1 (The govt. is picking up $5.50 and bidders are in for $1)

    At an avg. bid of $4 per hour, avg. worker is making $8, and the govt. is spending $250B a year.

    There is no more UI. There is no more minimum wage. That’s why there are 30M in program.

    ———–

    The goal here is to put the greed of private bidders to work to quickly identify which of the 30M are best as the jobs they are put too. When a family hires a babysitter, and after 6 weeks they have consistently rebid, that is known, expect the market to quickly clear to find the correct price of babysitters – trust the ones who clean to earn more.

    The goal is a fuzzy system, so a neighborhood with 30 dog owners can bid $3 an hour for someone to clean up their yards, and the guy taking the job knows he can do all 30 yards in 15 hours, netting him $20 per hour for picking up dog shit.

    The goal here is transparency. Workers and Bidders are encouraged to take pictures and video, document before and after. Consistent bidders with plenty of previous happy workers carry far more weight in judging employees than those with spotty records. Remember plenty of bosses now are LIKED but they can’t afford to keep good workers – this is a big change.

    ———

    The upside is ENORMOUS.

    1) while the govt. can’t bid it can now hire low cost services organized by SMB that hire from this pool. Suddenly the cost of public goods (clean parks, subways, etc) drops dramatically – making the cost of ALL GOVT. cheaper. This alone likely covers the expenditure.

    2) Because GI recipients will disproportionately fall in certain areas, entrepreneurs in these markets suddenly have a labor advantage that is real. Businesses that want to hire need to move into these areas.

    3) The signalling to the unemployed is REAL. This is not the unemployed doing socially valuable shit that Dan thinks matters, this is 30M seeing WHAT pays what in a sustainable way.

    4) The cost of goods and services in poor areas drops dramatically. Suddenly ghettos are cleaned up, properties are improved. Daycare drops to $50 a week. EVERYONE IS WORKING.

    5) This program allows us to move Housing and Energy Assistance and Foodstamps everything into it (even healthcare). If you are working for a boss who thinks you worth the market rate, we’ll cover your nut. Otherwise, you are in trouble.

    6) This program gives us a clear path to extending the age of retirement, first by choice, and slowly by law. It lets us ask even the disabled to work from home. I can all but guarantee 500K work from home customer service reps. There will be ZERO employees no one bids $40 a week on.

    The basic issue is that a Guaranteed Income is a social construct. It is what we determine OWE one another for being American.

    Meanwhile the value of a person’s labor can never be defined by fiat. Our workers are being competed with by global forces. They aren’t worth what you want to pretend. Stop pretending and let the market clear.

  117. K writes:

    JW Mason,

    “it is probably true that while the ineffectiveness of policy today is exaggerated, the effectiveness of policy in the past was probably also exaggerated.”

    So assuming we revoke cash, and rates can go negative. What exactly is the limit to the amount of demand that can be generated by the CB? Lets say the Fed cuts the policy rate to -20%. How will this *not* result in massive attempts at consumption/investment (much of which would be frustrated by inflation, of course)? And do you doubt that demand can be suppressed with excessively high rates?

    Or am I misunderstanding your point?

  118. Bryan Willman writes:

    On reflection, I think we (me included) have kind of missed a big related issue.

    ANY scheme to stimulate recovery from the recession faster than the current slow glacial pace is SOME KIND of allocation of policy-made luck to some set of winners, or allocation of policy-made misfortune to some pool of losers. In short, a forcing of the changing of the balance of holdings in the economy.

    Whether change in balance is good or bad policy for growing the economy is a topic of much debate.

    But the reality that it will *screw somebody somehow* is not in question.

    Inflate away debt – screws all claim-holders, which is everybody with at least $1 to their name. Net debt holders might get a win, for a while. But it’s hard to see how it’s a long win.

    Negative interest rates – will imply things like negative interest rates on checking accounts, and is not really different from inflation – screws everybody with any claim.

    Create artificial demand? Amounts to some variant of cronyism.

    It’s not a zero sum game, but it is played with a zero sum score card.

    [I am personally of the view that large natural forces of human behavoir surround debt, and the pace of recovery will be limited by those issues regardless of policy. But that's just saying we're screwed in additional ways beyond the one Steve points out.]

  119. Morgan Warstler

    Morgan, I love this idea, I’m just trying to work through it. I mean, the problem isn’t that there aren’t jobs, it’s that there aren’t jobs that pay well enough for people to take, unless they are ‘subsidized’ by cheap credit.

    How would people game the system?

    How do we regulate work performed such that the people offering work actually want the people willing to bid?

    I’m just wondering if this is a 1% solution or a 20% or what?

    Thanks

  120. @paul
    Youre half hearted apology is accepted.
    You’re lack of argument stands on its own.
    Thanks for the ad hominems
    Cheers.

  121. JW Mason writes:

    So assuming we revoke cash, and rates can go negative. What exactly is the limit to the amount of demand that can be generated by the CB? Lets say the Fed cuts the policy rate to -20%. How will this *not* result in massive attempts at consumption/investment (much of which would be frustrated by inflation, of course)?

    Well, first of all, we do not need to revoke cash to set a negative rate. Rates can already go as negative as the carrying costs of cash, which are not zero. But on the substantive question, a policy rate of -20% would translate into market rates in very uneven and unpredictable ways. Some would fall by a lot, some by a little. Some might rise. To the extent we got inflation, some interest rates are stickier than others, and some prices of course are stickier than others, so that would add to the unevenness of the effect. in short, we could not have the sort of ultra-expansionary policy you are describing without a massive and unpredictable alteration
    of relative prices whose effect on the real economy is unknowable and probably quite destructive. In practice, it would be absolutely impossible to pursue this course without a comprehensive system of price controls.

    Simple formal models are fine within a familiar domain where we have plenty of real-world experience to hone our intuitions. When you are considering a major departure like this, you have to think more concretely.

    My point is not that more expansionary policy is not possible, but that more expansionary policy is not just a matter of turning a single knob labeled “Fed Funds rate” (or “reserve quantity”).

    And do you doubt that demand can be suppressed with excessively high rates?

    To be honest, I have some doubts here too. First, increases in the policy rate do not always translate into increases in market rates. And second, the share of the economy accounted for by interest rate-sensitive sectors varies historically. But yes, in general, it is easier to pull a string than to push it.

  122. Dan Kervick writes:

    Since you are only talking to Morgan now, let me answer my own question. The implied asset in your example, of course, is *reserves* which cost $50,000 to hold.

    No, K, if you are talking about examples 3 and 4, then remember that in those examples I was assuming that there is no interest on reserves – not positive interest or negative interest. Instead, there is only a negative interest rate on discount window borrowing. Thus, the transaction goes like this: the bank borrows a million dollars in reserves from from the Fed on Monday, at -5% interest, and the Fed credits the million dollars to the bank’s reserve account. As a result, the bank owes the Fed $950,000. On Tuesday, the bank makes that $950,000 payment to the Fed and the Fed debits $950,000 for the bank’s reserve account. The business between the Fed and the bank is now concluded, and the bank has $50,000 in additional reserves.

    This, not coincidentally, leaves the bank with a profit of zero. If they had lent the money to a client at -4%, the profit would have been $10,000.

    The first sentence is incorrect for the reason I just stated. But even if it were true – as it would be in a situation is which the imposes negative interest on reserves, a “holding fee” so that the profit from the -5% borrowing is wiped out by the holding fee – the second statement is clearly wrong. The bank doesn’t “lend the reserves”. The reserve transaction and the lending transaction are two operationally distinct and independent operations. If the net impact on the bank’s balance sheet from the transaction with the Fed is X$, and the net impact on the transaction with the loan customer is $Y, then the net impact on the bank of both transactions together is $(X+Y). The net impact of making a -4% loan of $1 million is clearly -$40,000. So if you are assuming the net impact of the bank’s business with the Fed is $0, then the total impact is -$40,000.

    In my third and fourth example, I had assumed no IOR and a -5% borrowing rate at the discount window – in other words, the Fed pays banks to borrow and charges no “holding fee”. That scenario should be more favorable to your story, if anything, because at least in this case, if the bank borrows at -5% and lends at -4%, it does make a $10,000 profit. But what I argued is that in that case, the bank still has no incentive to make the -4% loan, because if it borrows the $1 million and does not make the loan, it nets $50,000 instead of $10,000.

    Remember, you are assuming no physical cash. The bank can’t somehow withdraw its reserve balance and give it to the loan customer. Its business with the customer consists entirely in debits and credits to the customer’s account. Similarly the Fed’s business with the bank consists entirely in credits and debits to the bank’s reserve account.

    In every case you failed to ensure that there were the same quantity of assets and liabilities. It’s the same thing that you did in comment 53, when you said that a bank that gives a loan to a customer who transfers the money to JP Morgan doesn’t have to raise funds to cover the loan, and instead just has to ensure that it has adequate capital..

    You are forgetting the asset represented by the loan itself, K. A deposit balance at a commercial bank is a liability of the bank; it is a debt the bank owes the depositor. If a bank extends a loan to a customer and thereby creates a deposit account for that customer credited for the amount of the loan, then the bank has acquired a liability and the customer has acquired an exactly equal asset. But of course, banks don’t just go around creating deposit balances for people for nothing. They exchange those balances for repayment promises. At the same time that the bank creates the deposit balance for the customer, the customer signs a piece of paper promising to pay the bank the same amount, plus interest. That piece of paper is the proof of a legally binding commitment and is an asset of the bank and a liability of the customer.

    So, if my bank loans me $5000, and I promise to pay it back by this time next year with 10% interest. Then my bank now possesses both a liability and an offsetting asset; and I possess an asset and an offsetting liability. My bank has made a promise to pay me $5000 on demand, and I have made a promise to pay the bank $5500 by next year. Apparently, the swap of promises is worthwhile for both of us due to time preference. I prefer $5000 now to $5500 next year, and the bank prefers $5500 next year to $5000 now.

    That’s where the corresponding asset comes from to match the liability the bank created – from the repayment promises. When the bank examiner comes around, the examiner counts all of those promises the bank’s customers have made as assets of the bank. The bank does have to possess some actual capital as well, because the examiner rates those assets in terms of their relative degree of risk. But the capital requirement is not anywhere close to the sum total of the bank’s liabilities. The risk in effect discounts the value of the loan repayment promises on the bank’s balance sheet, so some capital is needed in order for the bank to have positive equity. The bank does not have to fund its lending by acquiring an additional asset equal to the asset it acquires in the form of the repayment promise.

    On the payment to Goldman Sachs, there is a reason that reserve ratios are so low. In the ordinary course of business most of the deposit balances at the bank are not changing much. The total volume of daily transactions on the deposit accounts is way less than the total sum of the deposit balances. So even if one customer immediately writes a check for the total sum of some new loan, that’s just a drop in the bucket. The bank might have to borrow additional reserves if it expects the volume of daily payments to increase, but the borrowing cost is a fraction of the amount of the additional payment volume. And even if the bank fails to borrow sufficient reserves and overdraws its reserve account, the Fed just credits the necessary amount to the bank’s account and charges a fee – which as I understand it is the discount rate plus a 1% penalty.

    Since you are only talking to Morgan now, let me answer my own question.

    For Pete’s sake, I blew off Morgan before to argue with you and pissed him off. Now I say one sentence to Morgan and you say I “am only talking to Morgan now.” This is worse than a high school dance :)

  123. Morgan Warstler writes:

    Curt, this is the full boat solution.

    “I mean, the problem isn’t that there aren’t jobs, it’s that there aren’t jobs that pay well enough for people to take, unless they are ’subsidized’ by cheap credit.”

    I disagree slightly here. This isn’t a problem. There are plenty of jobs representing the wants and desires of people who at a certain price they would pay for them.

    And without any social commitment, those jobs would be filled. But we’d have people at the low end (bottom 20-40%) living lives that simply are not acceptable to everyone else.

    It would look a lot like some Hispanics in America today.

    Because of the incidence of illegal immigration or because they are a more moral people, Hispanics are LOATHE to not make rent. So you see far more doubling up in households. There would be a lot more of this, if we ran a perfect market.

    This is actually another upside of my plan, it SOLVES the illegal immigration issue.

    With 30M US citizens immediately put to work:

    1. In many jobs, it closes the market to illegal immigrants. Hiring the US citizen is just cheaper.

    2. It proves to us culturally what, if any, jobs are simply beyond the pale for US citizens. This is field work. In a perfect market based system, the wages paid to find field workers would have to increase until someone chose to do the work. We don’t have a perfect market based system. Technically, the robot builders are being screwed by illegal migrant labor.

    So we aren’t in nay way subsidizing with cheap credit.

    Instead we have agreed to make a social compact that Americans deserve $X.

    Now that the tax-payers have funded this commitment, we have 15M weeks of human labor laying around, and we should be trying to recover revenue as possible on that expenditure.

    So auction it.

    ——

    No system is going to be perfect, but mine will get conservative support.

    No system is perfect, but mine will have the least amount of fraud. Fraud is caustic in a social safety net, it makes use not trust one another.

    15M unemployed and I assume anther 15M currently employed find their way into this system. I think they all end up working within a year.

  124. Morgan Warstler

    Morgan thanks for taking the time to answer me. Thanks for proposing a market oriented system.
    As someone with some knowledge and experience with the kind of technology, organization, people and processes you’re talking about, I’m just far more skeptical than you are that it wouldn’t devolve into a 1% solution.

    That doesn’t mean I don’t like it or appreciate it. It means that it would far more likely result in being successful in some very small market. It would develop a reputation for that kind of market. That kind of market would represent a signal. That signal would not be ‘good’, and the remaining population would avoid it rather than signal it.

    I mean, the individual states have had this kind of thing for decades. At least since the late 70′s. It’s just a gutter labor pool for the incredibly desperate.

    Technically it’s trivial until the government gets involved. And in my experience the kind of people who work on these projects in government are one or two standard deviations below that of their private sector counterparts. The rules would become unmanageable. … The more I think about it the more exasperating it seems. The business owners would have to find some way of qualifying and screening applicants. At the bottom end of the pool the big problem is criminal records, drug use, an behavior. None of which come across online. … I just dont’ see how to make that work. There is a much easier version already: drive to your home depot. Negotiate for somewhere between 5 and 12 bucks an hour, feed the guys lunch… that’s how it’s done today. I’ve done it. It’s how labor and construction are done. And you can SEE the people. So what’s better about your system than that and craigslist? I just don’t get it. Now, if you mean government labor, then I’m all for it. But it won’t work in the private sector, and the costs of administering labor directly versus through a contractor who already knows his local labor pool is just nuts. So again, i just feel like you’re applying a middle class technology for the educated, to a lower class problem for the not.

    Love it. But just don’t see it working. thanks. -Curt

    But if you can push it, I’d be thrilled if you could get it off the ground.

  125. [...] — first Japan, now the United States and Europe — are obvious to a dispassionate observer. Their overwhelming priority is to protect the purchasing power of incumbent creditors. That’s it. That’s everything. All other considerations are [...]

  126. Morgan Warstler writes:

    “I just don’t see how to make that work. There is a much easier version already: drive to your home depot. Negotiate for somewhere between 5 and 12 bucks an hour, feed the guys lunch… that’s how it’s done today. I’ve done it. It’s how labor and construction are done. And you can SEE the people. So what’s better about your system than that and craigslist?”

    To start with, Craigslist is anonymous.

    This is EBAY reputations + real names and faces. If you always deliver as promised as buyer or seller, your reputation improves, and the amounts bid increase. Folks with bad feedback stop selling because they can’t get good prices.

    BUT no one can quit this system, not if they want to eat. AND they can be thrown out. Yes, since this is a safety net, throwing someone out would be a high hurdle, but at least that’s what we’d be talking about later. It is a good question to ask.

    i suggest the true droth ends up being punted further down worker chain until they are sitting at home being forced to keep answering the phone 40 hours a week for $6.50 a hour. Now TM companies can take the recorded calls, proving the worker is not even trying to sell home repair in the ghetto for new low cost construction.

    My point here is, we can’t make someone who is determined not to try, try – BUT we can torment the living shit out of them and feel ok about that part of the system. It’s the rubber room for child molesting teachers EXCEPT they get paid $6.50 an hour and are miserable with the phone ringing all day.

    To better shape this story, I’d also tel you to this gains from being a LOCAL services biz, like Yelp and Angieslist, but for not just for Home Depot labor, also for babysitters, house cleaners, yard workers, dog walkers, cooks, laundresses, etc.

    ——

    Imagine that you can now staff up a whole truck load of workers for your winning bid of $4 per hour, and spend all day fixing up distressed properties to flip homes. AND IT IS LEGAL.

    Think about the entrepreneurial drug seller in the housing project who now can buy up half the buildings residents for $2 per hour, and send them out doing legal service jobs.

    Establish a reward for proving that people aren’t working and are trying to kick back to a fake employer who bids $1 (harsher penalties for him), and low cost enforcement kicks in as profitable venture.

  127. econ writes:

    You need to read up on Arrow’s Impossibility Theorem (which was his dissertation, by the way). You say that the preferences of polities are “rational.” This may be true of, say, Syria and North Korea, but as Arrow proved, democracies without dictators cannot have rational preferences. Just isn’t possible. Polities either have incomplete preferences, intransitive preferences, or dictators. Since democracies don’t have dictators, the preferences cannot be rational. Sorry.

  128. JW Mason writes:

    Dan,

    So what about the $1 trillion in excess reserves currently being held by banks? A tax on excess reserves looks, formally, a lot like a negative policy rate. So the question is, can banks respond to such a tax without increasing the supply of credit to the real economy? I admit, I’m genuinely not sure of the answer.

  129. K writes:

    Dan,

    I hadn’t noticed you went back to zero IOR in those examples. That just makes no sense at all. If banks can borrow reserves at -5% and hold those reserves at zero, then they will do that in unlimited quantities and make unbounded profits. Reserve interest has to be less than or equal to FF which in turn must be less than or equal to the discount window, or he whole system fails. Let’s not discuss absurd systems.

    “You are forgetting the asset represented by the loan itself, K”

    No, never. Where?

    You said banks wouldn’t lend at negative rates. My challenge to you in comment 73 was to show me why they wouldn’t in the system I outlined which is exactly the system we are in right now except without cash and with negative FF. I said the reserve rate was equal to the FF rate. You tried that in your first two examples but those were incorrect. Your second two examples weren’t the system I proposed (but also don’t make sense). The system I showed you works exactly as the same as our current system but with negative rates. If you can’t demonstrate why it wouldn’t work, I consider our disagreement settled. Does that not make sense?

    “This is worse than a high school dance” :-)))

  130. K writes:

    JW Mason,

    If they impose significantly negative IOR, banks will just exchange excess reserves for paper money. That’s their right.

  131. K writes:

    JW Mason,

    Sorry, were you talking about a cashless economy? Then, assuming FF is equal to the reserve rate, all risk free asset rates will go negative and the effect will propagate through the economy. What would stop it?

  132. Morgan Warstler writes:

    Let’s be clear here https://twitter.com/#!/interfluidity/status/193111787615424514

    Steve calls my idea interesting…. Dan HIDES.

    wimp.

    I sell my idea as hereabouts the thing that shuts up MMT. In one move. Checkmate. Smell the glove MMTers.

    Do I rock or what? Wait till I sell it to the Tea Party….

  133. Morgan Warstler writes:

    1. First prove I silence MMT on JG.

    2. Carry the win into normalized salons.

    3. Profit? Don’t you love it when the ? come last?

  134. Morgan Warstler writes:
  135. Dan Kervick writes:

    OK, K, so I guess at least you are no longer saying that banks need to fund their loans by attracting additional deposits or raising additional capital equal to the amount loaned.

    So let me understand what we are supposed to be imagining. There is a negative FF rate, so the bank makes money from borrowing. But there is a negative IOR rate, so the bank loses money from holding any funds it has borrowed. How big is the gap? If a bank borrows $1 million and holds it for one month, does it make money or lose money?

  136. Dan Kervick writes:

    Morgan, so does your system essentially amount to eliminating the minimum wage but setting a minimum income, and then having the government pay the worker the difference between the wage the firm pays and the minimum income?

  137. K writes:

    Dan,

    “at least you are no longer saying that banks need to fund their loans by attracting additional deposits or raising additional capital equal to the amount loaned.”

    I never said they need to fund their loans with “capital.” If I ever say a bank needs more “capital” it means that they need to increase their ratio of junior to senior liabilities, like issuing equity to buy back debt. But I could easily say a bank “needs to fund its lending with deposits or borrowing.”

    I’ll tell you the difference between you and me: When you say “a bank lends $1 million,” to you that means that bank also gets $1 million of deposit funding. When I say the same thing, my default assumption is that the associated deposits end up on other banks’ balance sheets. The bank therefore has to borrow that money, for example in the interbank market. So when I say “a bank has to fund its lending,” you think I don’t understand that the lending creates deposits. *I do. I really do.* It’s just that they are some other bank’s deposits. And when you say “A bank doesn’t have to fund it’s loans,” I think “but the deposits are somewhere else. Dan must think that a bank can have assets without matching liabilities.” I don’t think we disagree on accounting (but I do think we might disagree on the economics of the bank’s lending decision). Does that make sense?

    “If a bank borrows $1 million and holds it for one month, does it make money or lose money?”

    In a typical IOR system, there’s an “operating band”: The interbank rate is above the reserve rate and below the discount window. That keeps the interbank market working by insuring that lenders prefer interbank over reserves and borrowers prefer it over the discount window. In Canada I think the band (discount rate minus reserve rate) is 25 bps. For our purposes, I think there’s no harm in assuming that all three rates are the same. So your hypothetical bank loses a bit of money in the real world, but in our discussion I think we should assume it breaks even.

  138. [...] in the comments at Steve Waldman’s, Morgan Warstler has an interesting proposal for a replacement for the minimum wage and unemployment insurance: Using a clone of Paypal and Ebay [...]

  139. paul writes:

    @Curt

    “Youre half hearted apology is accepted.” – Curt

    It wasn’t an apology.

    “You’re lack of argument stands on its own.” – Curt

    I made counter-claims to your comments. You didn’t offer any meaningful response. No argument was allowed to develop.

    “Thanks for the ad hominems.” – Curt

    No ad-hominems in my original comment. True i was rude when I normally wouldn’t be but that’s mood swings and the internet. Still not apologizing.

    Don’t see any ad-homs in my follow-up either. Just a lack of respect for your worldview that puts producers above workers and tends to discredit those that don’t compete well (enough) for the producers as lazy.

    If the state didn’t monetize production there wouldn’t be one-tenth as much there is. You seem to be oblivious of that.

    You still haven’t written anything of substance. Rather you have responded to the tone of my comment and not the comment itself.

    Your responses have been a collection of talking points. Maybe from Mises.org?

  140. Morgan Warstler writes:

    “Morgan, so does your system essentially amount to eliminating the minimum wage but setting a minimum income, and then having the government pay the worker the difference between the wage the firm pays and the minimum income?”

    Dan, please read I am claiming many, many advantages. The first of which is that the state cannot direct the labor of the unemployed. Everyone is auctioned, so they have a bossy boss who has paid $ for their labor and directs them.

    In my system Dan is powerless.

    I also solve for illegal immigration, dramatically drive down the cost of public employees, enable the seamless increases of the retirement age, and fix ghettos.

    And MINE CAN PASS MUSTER with conservatives.

    Mosler leads a band of dirty hippies, and you aren’t making him look good.

  141. JW Mason writes:

    If they impose significantly negative IOR, banks will just exchange excess reserves for paper money.

    Only assuming that cash has zero carrying costs. But this is not true, it has significant carrying costs. We know this is true because taxes on reserves have actually been adopted in Sweden, and seriously contemplated in the UK and elsewhere, without any elimination of cash, and because we have observed negative interest rates on short-term government bonds in the wild. It seems likely that the costs of holding cash are high enough to leave space for a couple points of negative IOR. And if we did reach a point where the conversion of reserves to cash became a constraint, we would not have to abolish cash to eliminate the constraint. The central bank could just start exchanging reserves for cash at less than par. I don’t know if there are legal obstacles to this, but there is certainly is no technical obstacle.

    So yes, let’s say that for whatever reason banks do not or cannot exchange excess reserves for paper money. How concretely does this translate into increased credit availability and activity in the real economy? I’m not saying it doesn’t, I just want to be sure I understand the mechanism.

  142. Peter K. writes:

    studentee

    “Dan Kervick usually brings something originally and thoughtful to every discussion I’ve seen him take part in. You on the other hand, just seem to follow him around, dropping awful, consensus-mongering prose. ”

    Your comments are the worst also. The Internet has given a forum to some really bent views. They post a lot because it’s their only outlet and the effect is that certain views are given more weight than they deserve. It’s just some cranks posting a lot.

    I generally agree with the bloggers where I comment. Waldman, Baker, DeLong, Krugman, Yglesias, Thoma, etc. Kervick doesn’t. It’s Kervick who is intruding with muddled, nonsensical arguments wasting people’s time and energy. Krugman tried to engage with the heretical Naked Capitalism blog – probably b/c he is often treated as heretical – then was sorry he did. If it weren’t for Naked Capitalism would we even be talking about monetary mysticisicm.

  143. paul writes:

    “The Internet has given a forum to some really bent views”

    Automatically self-annointing your own views with mainstream Holy Water by associating yourself with the mainsream views that have done a very poor job of informing successful policy outcomes.

    Your views are the most bent because they are impervious to alternate thinking based on logic and fundamental arithmetic.

    You are a follower. Part of the problem not the solution. You know the definition of insanity…doing the same things over and over expecting different results.

    I would be troubled if you agreed with me. Dan at least thinks for himself.

  144. JW Mason writes:

    Having finally skimmed this long thread (perhaps not the best use of my time), I think that K is basically right and Dan Kervick basically mistaken about the logic of negative rates.

    However, I think that while K is right at the level of abstraction of this conversation, he/she is ignoring some concrete issues that arise with negative rates. In particular, even if we eliminate cash, it is not follow that there is no zero-return instrument available, and that the whole complex of returns could just shift smoothly downward. In the first place, there are many standard financial contracts which implicitly include a zero interest rate along with other terms. These are not going to disappear or be rewritten to implicitly include an interest rate in the face of negative IOR. Similarly, the types of activities that can generate a zero return are different from the types of activity that generate a positive return. If you are looking for a borrower who can pay a positive interest rate, you generally need to look for someone engaged in productive activity of some kind. If you are looking for a borrower who can pay a zero percent interest rate, it’s enough to find someone holding a stock of some physical asset with a modest carrying cost. And the lower the policy rate, the longer the list of physical assets with low enough carrying costs to make positions in them worth financing.

    In short, while you are right that formally, a negative IOR plus limits on banks ability to convert reserves into cash would induce additional lending, I think it’s dangerous to assume that this new lending would be qualitatively similar to what happens when the policy rate falls to a lower but still positive value.

  145. JW Mason writes:

    (Again, I agree with SRW that depression is a choice. But I don’t think there’s a single easily tweaked monetary policy tool that would allow us to chose not-depression. It will take some “bold persistent experimentation” with a variety of policy and institutional changes.)

  146. Dan Kervick writes:

    I’ll tell you the difference between you and me: When you say “a bank lends $1 million,” to you that means that bank also gets $1 million of deposit funding.

    I’m sorry but I have no idea what you mean K. I think in many cases the bank gets no additional funding. It creates the deposit balance for the customer out of thin air. The folks at the bank whose job it is to manage liquidity are making sure that the bank’s total reserves are adequate to make its estimated payments given its total volume of deposits, so if the deposits grow by a certain amount, they may have to acquire additional reserves – but generally that will be a fraction of the amount of growth of the deposit balances.

    I’m also puzzled by the idea of a bank “having assets without matching liabilities” which you say I am assuming. Which assets are you talking about? The deposits are liabilities. Those liabilities might be created by – for example – a walk-in deposit. In that case the depositor gives the bank some cash, which becomes part of the bank’s total reserves and a bank asset, and in exchange the bank creates a balance in a deposit account for the customer, which is part of the bank’ liabilities. But the deposit account balance might come into existence as a result of a loan. In that case, nobody deposits any cash (or writes a check or wires a payment etc.) that is “funding” the loan. The bank creates the deposit balance, and in return they receive a promise of future payment from the loan customer. That promise is the asset they acquire in exchange for creating a liability.

    Also, businesses can frequently have assets without matching liabilities. In fact they all want to have assets exceeding their liabilities. In the financial system as a whole assets and liabilities must be equal. One business’s asset is someone else’s liability, and vice versa. But inside the individual firm assets and liabilities need not match.

    Now back to the point about the cost of borrowing, if we are assuming that both the FF rate and IOR are adjusted so that while both are now negative, the bank still either loses a bit of money or breaks even on borrowing, then I’m mystified about what the purpose of the hypothesized negative rate regime is supposed to be. The real cost of borrowing hasn’t changed. Wasn’t Yglesias’s point supposed to be that the zero bound currently represents a boundary below which the Fed cannot push the real cost of borrowing, and that if it could puncture through the zero bound it could reduce the real cost of borrowing? But if the Fed only changes from a current system in which it charges banks money to borrow reserves but then pays most or all of that money back to them via IOR to an alternative system in which it pays banks money to borrow reserves but then taxes most or all of that money back through negative interest in reserves, then what has actually changed? The real cost of borrowing is the same.

  147. Dan Kervick writes:

    J.W. Mason said:

    These are not going to disappear or be rewritten to implicitly include an interest rate in the face of negative IOR.

    Yes, this seems to me to be the crux of the matter. People will want to save no matter what. If somehow the government turns dollars into a vehicle whose nominal value continually depreciates as it is being held, people will seek to convert that vehicle into something else. Financial value is ultimately grounded in the value that is being added to real goods by economic production. If the dollar is turned into a deteriorating claim on real value, people will fly into some other kind of financial instrument, which innovators will be happy to invent.

    In short, while you are right that formally, a negative IOR plus limits on banks ability to convert reserves into cash would induce additional lending …

    The thing I’ve been having trouble with here is the notion that this change in IOR would increase the bank’s real cost of borrowing reserves. If the the negative IOR is combined with a corresponding negative rate for borrowing reserves – either via interbank borrowing or discount window borrowing – then the real cost of borrowing hasn’t changed.

    So what about the $1 trillion in excess reserves currently being held by banks? A tax on excess reserves looks, formally, a lot like a negative policy rate. So the question is, can banks respond to such a tax without increasing the supply of credit to the real economy? I admit, I’m genuinely not sure of the answer.

  148. K writes:

    JW Mason,

    [Response to your comment 141. Haven't read the rest yet.]

    OK, fine. Cash has some carrying cost. And it takes a few days to get it so unless you have a permanent positive excess reserve balance, you are not going to have time to convert it to cash. Anyways, to seriously contemplate ending the liquidity trap, you need to be able to go below -1% or whatever, or you are just faced with a slightly different lower bound. If the FOMC sets fed funds at -5% (with a necessarily equal or lower IOR) for 6 weeks, you will not end up with $3 trillion of reserves, or whatever we’re currently at. Instead, the Fed will get lifted at -4.5% at the discount window (the interbank market will stop immediately) in unlimited quantities and there will be equally massive requests for $1000 bills. The system will blow up. You can only go negative *as a matter of policy* by eliminating cash. (In fact, you could just *restrict* the quantity of cash. That will permit a limited arbitrage trade, but paper money will disappear from circulation.) One things for sure: bank lending rates cannot drop below zero as long as borrowers can stash away the borrowings in cash. That’s a real floor on the demand stimulus potential.

    “let’s say that for whatever reason banks do not or cannot exchange excess reserves for paper money.”

    It’s the funds rate that impacts the economy. Lets say there is no cash. If you put fed funds at -5% and IOR at -10%, its the -5% rate that effects bank lending rates. But there will be *zero* excess reserves. To do QE (and get excess reserves) you need the reserve rate to be effectively the same as the fed funds rate or banks will dump those reserves in the interbank market until the fed funds rate is equal to the reserve rate. You cannot have any significant quantity of outstanding excess reserves at a rate that is significantly less than the fed funds rate.

  149. K writes:

    JW Mason,

    [Next comment...]

    “I think it’s dangerous to assume that this new lending would be qualitatively similar to what happens when the policy rate falls to a lower but still positive value.”

    So there’s something special about nominal zero? You are talking about dropping risk free real returns, below the real return on real assets, right? (That’s the whole *point* of stimulus). If inflation was at 10% instead of 2% then would you say there is something qualitatively different about going below 8%, or do you still say 0% nominal is something special? To me, there’s a natural rate. If it pukes because of a severe shock, then you need to get the real rate below it. The nominal rate is irrelevant.

  150. K writes:

    Dan,

    Equity is a liability. In the case of changes in asset value, the equity value is what changes to make assets and liabilities the same. No?

    Lets take a virgin bank. It has $1 million of equity and $1 million of reserves. Nothing else. Then it makes a $9 million loan to me. I transfer the money to JP Morgan. What, in you opinion, are the assets and
    liabilities on the banks balance sheet after these initial transactions are done? What is the accounting value of each of these assets and liabilities (including equity).

    “The real cost of borrowing is the same.”

    What I demonstrated above is that nominal lending rates will follow the interbank rate down below zero. A -5% interbank rate will result in, e.g. a -4% lending rate. Why do you say real costs of borrowing haven’t changed?

  151. Can creditors be better off in a stable economy? Like the tootsie roll pop, the world may never know. Nice read. Thanks.

  152. [...] grotere schaal afschrijven van schulden wellicht ook een optie is om de Euro te laten overleven vindt u hier. Het is nu nog een stap te ver, voor Klaas. Maar hij komt er [...]

  153. Dan Kervick writes:

    What I demonstrated above is that nominal lending rates will follow the interbank rate down below zero. A -5% interbank rate will result in, e.g. a -4% lending rate. Why do you say real costs of borrowing haven’t changed?

    The assumptions are that nobody can hold physical cash, and IOR is the same as the FF rate. For the bank to borrow it has to borrow FF into its reserve account at -5%, and the Fed then imposes IOR of -5%. As you said, that means the bank breaks even. So from the bank’s point of view it doesn’t matter whether the FF rate and IOR are both -5% or whether they are both 5%, or 10% or 20%. They break even in all of those cases. This is just an extension of the current ZIRP policy.

    So what about the loan customer? I suggested earlier that the bank would never have an incentive to offer a zero nominal rate loan, because its lending operations with customers and reserve operations with the Fed are operationally independent. So no matter what choices it makes on the reserves side, it only degrades its balance sheet by making a zero interest loan, since it receives less money back than it lends out. You said in reply that the interest rate on the borrower’s account would also be negative, since we should assume all interest rates are negative. Although I can’t recall if you specified some particular number, I assume you are saying that the situation with the loan customer is parallel to the situation with the bank: the borrowing rate and the holding rate are equivalent: just as the bank can’t make money from the Fed by borrowing, so the loan customer can’t make money from the bank by borrowing. The bank doesn’t lose by lending because in the cashless society the borrower has to hold the cash as an account balance, which earns negative interest from the bank, thus cancelling out the gain the borrower made by borrowing the money at a negative loan rate.

    OK, fair enough. But now the supposed benefits of this negative interest rate policy in a cashless society have evaporated. If the only form in which I can possess money is as a balance in a bank account, then as far as my real cost of borrowing goes, it makes no difference whether I borrow at a positive rate of interest and then receive an equivalent positive return for holding the money in an account, or whether instead I borrow at a negative rate of interest but am then charged an equivalent negative return for holding the cash. The cost/benefit is the same for me.

    You might say that every one will seek to come out ahead by spending the money instead of holding it. But the macroeconomy as a whole can’t come out ahead in this way. For every expenditure there is a receipt; the increased readiness of the money holder to spend money in exchange for some good whose value is not declining as rapidly is offset by the decreased readiness of the possessors of goods to accept money. If all money exists only in the form of deposit balances, then spending in this economy consist only in deposit balances being credited and debited in equal amounts across the economy. And if every such balance is diminished in nominal value during each moment of its existence, then the overall demand for that particular form of asset – deposit balances, is going to decrease.

    As I said before, the result strikes me as massively deflationary – since the quantity of money continuously shrinks, prices universally start heading south.

  154. Jesse writes:

    I could not agree more, that Depression is a policy choice.

    The question is ‘who is we?’ Who is making that decision and why?

    Hint: I don’t believe it is ‘we,’ as in ‘we the people.’

    Thanks for your blog.

  155. Steve Roth writes:

    @Morgan and Curt:

    I like the idea, Morgan, and Curt I get the issues you raise.

    Which is why I tend to rant, instead, for a major expansion (and simplification) of EITC, delivered farther up the income spectrum and with a slower ramp-off (to reduce marginality issues), with the credit delivered on weekly paychecks in the same way that taxes are deducted.

    The auction still exists out there, but without a central clearinghouse.

  156. [...] Yglesias adds his take HERE: This makes me think that explanations for the poor performance of macroeconomic stabilization policy that focus on the interplay between debtors and creditors may be reaching for something too [...]

  157. Max writes:

    “(In fact, you could just *restrict* the quantity of cash. That will permit a limited arbitrage trade, but paper money will disappear from circulation.)”

    Definitely not, since it still fills a need. And the paper, although not money, would be far closer to money than any physical alternative (foreign currency or precious metals).

  158. K writes:

    Dan,

    Lets say inflation is running at 2% and I have project that I expect to deliver a *real* return of 2% over the period of one year, equal to a 4% nominal return. Assume that I expect a 12% real return if the economy goes well, but a -8% real return if it goes badly, with equal probability. I’m very averse to that kind of market risk, so I wont do the project unless I can expect a profit of 5% (that’s the risk premium of the asset). The fed funds rate is 0%, and I’m an excellent credit so I can borrow from the bank at 1%. If I borrow at 1%, the expected return on my investment is 4%-1%=3%, which is not enough to compensate me for the risk. So I wont do it.

    Now imagine that the Fed cuts the policy rate to -2%. My borrowing rate will be -1% and the expected return on my project is 4%- (-1%) = 5% which clears my required expected profit. So at any policy rate below -2% I do the project. Stimulus accomplished.

  159. K writes:

    Dan,

    “it makes no difference whether I borrow at a positive rate of interest and then receive an equivalent positive return for holding the money in an account, or whether instead I borrow at a negative rate of interest but am then charged an equivalent negative return for holding the cash.”

    Your argument, if it were true, would apply equally to an IOR economy above or below the ZLB. You are saying that if we pay IOR at the fed funds rate and deposits earn roughly IOR, then monetary policy doesn’t work, either above or below the zero bound. But there are already plenty of economies like that. It’s called an “operating band” system. You seem to assume that monetary policy works by adjusting the difference between risk free borrowing costs and risk free deposit costs. That’s definitely not the case. It works by directly adjusting the real rate of interest relative to the natural rate, i.e. the required real return on investments (so do the other systems, but it’s less obvious, especially to monetarists).

    Did you think about my bank accounting example in comment 150?

  160. Feed Purge writes:

    [...] Economic depression is a choice [...]

  161. [...] The current recession is not inevitable, but a choice made by the polity. Meaning one way to end it is to change peoples’ minds. [...]

  162. Dan Kervick writes:

    Now imagine that the Fed cuts the policy rate to -2%. My borrowing rate will be -1% and the expected return on my project is 4%- (-1%) = 5% which clears my required expected profit. So at any policy rate below -2% I do the project. Stimulus accomplished.

    Well, I feel like we’re going in circles here, K, because as you know I have rejected the suggestion that lowering the policy rate to -2% will cause banks to reduce the commercial lending rate to -1%. The fact that the central bank gives commercial banks money for free does not in any way mean that the commercial banks have an interest in giving their customers money for free.

    On comment 158, you might know that I tend to be extremely skeptical about the efficacy of central bank policy and the outsized role that is assigned to it in recent economic thought. But here’s the way I think it works, when it does work: In our centralized banking system, bank liabilities are liabilities for base money, the final means of payment in our financial system. The Fed is the monopoly supplier of that final means of payment, and administers the reserve accounts through which bank payments are made. As a result, banks that are not carrying excess reserves are required to obtain additional reserves when they expand their deposits via lending. (They are required by law to obtain them if there is a reserve requirement; they are required by prudence to obtain them if there is not a reserve requirement.) There is a price for obtaining reserves. I believe central bank policy has whatever effect is has solely by it’s ability to adjust that price. That’s the only real transmission mechanism. All of the other alleged central bank powers invoked by monetary policy enthusiasts are smoke and mirrors.

    I don’t believe that in any of the mechanisms you have proposed you have presented a convincing case in which a commercial bank improves its balance sheet position by making a negative interest loan. The loan desk through which the bank relates to customers is one operation, and the liquidity management desk through which the bank relates to the Fed and interbank lending market is another operation.

    It is not enough to find a case of a negative interest loan operation in which (loan operation + liquidity operation) = balance sheet improvement. What you need to present is a case in which (loan operation + liquidity operation) > liquidity operation.

  163. Disillusionedliberal writes:

    “I detest the preference for depression revealed by my polity.”

    Oh spare me the self-righteous, santimonious pandering. Perhaps if the left hadn’t failed miserably, completely, and spectacularly to make another choice appealing and realistic I would sympathize, but our failure in that regard is no excuse for feeling sorry for ourselves. If there is “misery provoked by precarity and unemployment” in America it is because liberals failed, not because conservatives triumphed. Spare me!

    Or maybe you simply overestimate both the misery and the depth of empathy and compassion in the polity.

  164. Dan Kervick writes:

    I don’t know whether there is any kind of preference for depression economics in the public at large. People don’t really understand the latent power that is in the public hands, and don’t realize that there is a way out of the doldrums. The government – that is we, the public – can nuke the recession any time we want by first appropriating to ourselves as much purchasing power as we need and then by using that purchasing power with gusto. But ordinary folks don’t do public finance. The plutocracy, and the economists and political class who serve them, have bamboozled the public into believing that they are “out of money” – an inherently absurd idea in a fiat monetary system. The plutocrats are desperate to prevent a critical mass of voters from understanding that the dollar is a simple public monopoly that we legally control, and that we can bend to whatever purposes we want.

  165. Morgan Warstler writes:

    Steve, all the benefit is the central clearing house, it is there that you the positive and negative feedback loop 52x a year.

    Part of what my system does is quickly identify the real dregs, they get pushed to the bottom, they get given the $1 jobs where it is very easy to prove they are not working.

    And then BAM! They are suspended for 1 Month with no GI.

    That first month with no money off the dole, that might not do it, but on a three month suspension every Friday everybody else you know they have money on their debit card.

    You, your name and face, have a string of people showing pictures, audio, video of how you cleaned their garage, painted their fence, answered the phone. And it all looks like shit.

    BUT!

    Turning the corner is EASY. Do the first job offered. See there is proof! They hired me back! See there is proof.

    The feedback loop is sooooo constant, 52x a year. Good Job! Good Job! Bad Job. Good Job!

    This is far more than EITC or workfare etc. This is about SIGNALING, most people don’t get immediate knowable reward for doing better or punishment for doing worse.

    You can say that only a certain kind of guy can only live on commissions, but I know you force people to live on commissions and lot of people become that kind of guy.

    This is a system that changes bottom feeders. And it isn’t govt. doing it, it is the market doing it.

  166. Morgan Warstler writes:

    Dan,

    Let’s be honest here. Instead of being able to answer my awesome Guaranteed Income plan you are stuck, so instead you want to revert back to MMT chatter:

    “The government – that is we, the public – can nuke the recession any time we want by first appropriating to ourselves as much purchasing power as we need and then by using that purchasing power with gusto.”

    The govt. is not “we the people” not if you think the govt. has the power you you want to have, the last time we all said “we the people” it was said PRECISELY so guys who think like Dan don’t get to have any power.

    —–

    Look man, your inability to really answer my plan SPEAKS VOLUMES.

    I have bested you, silenced you, taken you to checkmate in just a few short moves.

    So you can fallback on mindless drivel about theoretical banking functions, but dearest Dan, right in front of you is a single policy proposal that ENDS UNEMPLOYMENT.

    And who needs MMT if there is no unemployment?

    And you can’t find a strong response. Mosler won’t have one either. None of you will.

  167. SKO writes:

    Forgive my ignorance, but I’m confused on a couple of points regarding the discussion between Dan and K.

    Wouldn’t a negative rate loan be an asset rather than a liability? Why would a bank need capital requirements at all to take out such loans? Even if capital requirements still bind, whats to stop a bank from taking massive loans from the CB in order to purchase any asset with a cost to carry less than the FF rate?

    Seems to me that would cause asset price inflation of an unprecedented scale since it would be basically impossible to lose when speculating. It’s not like you would be afraid of paying back your loans since they pay you. A loan with a negative real rate that results from inflation still leaves something on your balance sheet that is a net negative for you, even if its value is dropping, you still need to make the payments. With that reversed one should “borrow” to ones limit under any circumstance, no? Sure, if you left the balance in your account it would depreciate at a rate faster than the money comes in from bank, but why would that even bother you, hasn’t this process left you with a de facto asset?

    While I can see the mechanism that K is describing for why a bank would still make these loans, I’m not seeing why the bank’s borrowers wouldn’t immediately max out this offer. I’m probably missing something obvious, and I know this discussion has already overstayed its welcome, but I can’t help but be curious.

  168. Lenny writes:

    Steve Waldman wrote “We are choosing continued depression because we prefere it to the alternatives.” Are we bad to do that? Seems better than the destroying my savings. Is this just a blame game.

  169. K writes:

    Dan,

    ” I have rejected the suggestion that lowering the policy rate to -2% will cause banks to reduce the commercial lending rate to -1%”

    You tried to “reject” it and failed. Both of your examples had basic accounting errors. You rejected *nothing*.

    “It is not enough to find a case of a negative interest loan operation in which (loan operation + liquidity operation) = balance sheet improvement. What you need to present is a case in which (loan operation + liquidity operation) > liquidity operation.”

    Challenge accepted. What is the “liquidity operation” that generates equal or superior profit compared to the loan? Just *once* please give me something coherent. Every liquidity operation generates exactly zero profit, Dan. Go back to your comment 80, fix all the errors and show me how I’m wrong. Don’t forget: reserves, fed funds, discount window, and deposits at -5%; commercial and consumer loans at -4%.

    And here, so you don’t need to refer back to it, is the question that you persist in ignoring:

    “Lets take a virgin bank. It has $1 million of equity and $1 million of reserves. Nothing else. Then it makes a $9 million loan to me. I transfer the money to JP Morgan. What, in you opinion, are the assets and liabilities on the banks balance sheet after these initial transactions are done? What is the accounting value of each of these assets and liabilities (including equity).”

    Come on, Dan. It’s *trivial* for a banking expert. Just answer it. The answer could fit in a single tweet.

  170. K writes:

    SKO,

    “Wouldn’t a negative rate loan be an asset rather than a liability?”

    I give you a $1 million dollar one year loan at -1%. I should consider that a liability? I lent you *plus* one million.

    “I’m probably missing something obvious”

    You’ve missed the fact that there is no risk free nominal asset that earns zero percent. Don’t talk in vague generalities. What asset *exactly* do you want to buy with your -4% interest loan??? There’s no cash and deposits earn -5%.

    “I know this discussion has already overstayed its welcome”

    If so, I wish Steve would weigh in on one side or the other, and end it. It seems pretty critical to determine whether a proposal is actually possible before considering whether it would be good or bad, no?

  171. K writes:

    Lenny: “Seems better than the destroying my savings.”

    Then why don’t you buy some *real assets*, rather than sitting on your ass and depending on the government to insure your future consumption. Your consumption guarantee is backed by my taxes. Take some risk, you wimp.

  172. K writes:

    JW Mason,

    There may be nominal zero interest assets in the economy. That doesn’t cause any problems for negative rates any more than 5% coupon bonds cause a problem for 4% nominal rates. Cash prevents negative rates for the *only* reason that the fed will provide it in unlimited supply in exchange for reserves. So if you can acquire reserves at a rate below zero, there is an unlimited arbitrage available. And the reserve rate must be below the fed funds rate, so ff can’t be negative if there is cash around. But the fact that there are some nominal zero interest assets around just means that those assets will trade at a premium that depends on the zero coupon yield to the maturity of those assets. Currently such assets are worth less than par (unless they are putable). In a negative rate environment they be worth above par. And unlike cash, *nobody* is going to create them in unlimited supply.

  173. [...] Randy Waldman esittää omassa blogissaan hyvin, hyvin häiritsevän pohdinnan siitä, että lama saattaakin olla tietoinen valinta. Pelottava ajatus, joka ansaitsee harkintaa sekä hieman pohdintaa kotimaisesta vinkkelistä. [...]

  174. SKO writes:

    K,

    I have no argument with the idea that in the eyes of the bank making the loan it would be seen as an asset. I have a problem on the other side, from the borrowers perspective.

    I take out a million dollar loan at a rate of -2%. I will now receive $20,000 every year. If I then go and buy a 2% bond from someone, perhaps with the million dollar loan I took out, how is the bond I now possess in any way different from my loan? A regulator or accounting agency might declare that one is an asset and the other a liability but in de facto terms there appears to be no difference. Now presumably no-one would be crazy enough to offer perpetual loans under these circumstances but the result would be that repayment schedules would have to be fast enough to force net cash flows for the borrower to turn negative. Otherwise, there would be no reason not to maximize your debt load. I can’t help but think that the addition of this new institutional constraint would be highly disruptive.

    “Don’t talk in vague generalities. What asset *exactly* do you want to buy with your -4% interest loan??? There’s no cash and deposits earn -5%.”

    As JW suggested, I could purchase basically any asset with a carrying cost less than the deposit depreciation rate: stocks, forex, gold, houses, tulips… it doesn’t make any difference because there is no reason to expect any of these assets to suffer deflation, quite the opposite actually since everyone will be looking to convert their money into something that is not guaranteed to lose value. I wouldn’t even have to liquidate my assets when the loan repayments come due since it will be in my best interest to roll over the debt as long as rates remain negative. I just don’t see how cash substitutes wouldn’t arise in this situation.

  175. Trixie writes:

    Morgan,

    I think your e-bay hiring hall is a fantastic idea:

    http://www.youtube.com/watch?v=LJGlE0qA_rA

    Once? I poured hot coffee onto my crotch. Just because. Another time? I thought your paper towel dispensers were 2 inches too “high”:

    http://www.youtube.com/watch?v=cS8BNuTIlko

    Finally? I found your penis 2 inches too “low”. So, I cut it off and placed it in a jar of formaldehyde. You sued for unlawful possession, of all things. I won. Something about “squatters rights” by the time it came to trial.

    So good luck with your “umbrella” insurance. As Beowulf says “You can thank me ($$$) later…once you patent your idea.” He’s a lawyer after all, hopefully a personal injury one. It’s all about “dot-connection”, you know.

    And then remind me again who “runs” this country. Hint: It’s not about you and your “hard assets”, in the event you will have any left. Literally and figuratively, that is.

    All in, thanks for the $240/week to sit in my basement, smoking pot while laughing at you.

    Signed,
    Dirty Hippie

  176. JKH writes:

    Apart from the problem of currency, interest rates can operate at negative levels, in theory. The only critical requirement is that the CB retains control over the target rate for funds.

    This can be accomplished by setting IOR at no more than the floor for the target trading range for funds. That rule holds for corridor and floor systems, in positive and negative interest rate environments.

    E.g. if the corridor target is 2 and the floor is 1, the Fed must pay IOR at no more than the floor. Pre-2008, it paid IOR systematically at zero, which was always less than the corridor floor, because the zero bound was never an operational issue.

    Now, it’s paying IOR of 25 basis points, because it’s converted from a corridor system to a floor system, a necessity due to the level of excess reserves it’s created from various asset programs. It’s chosen to set the floor at 25 basis points rather than 0, for various operational reasons. Funds trade below the floor much of the time because certain depositories don’t earn IOR, and arbitrage is inefficient.

    E.g. if the floor is (3), the Fed must “pay” IOR at no more than the floor. It could “pay” (4) in a corridor system but would “pay” (3) in a floor system. So long as (target funds – IOR) is non-positive, this works whether rates are positive or negative.

    (The target pretty much collapses to the floor when the system is in a state of chronic oversupply of reserves, as is presently the case.)

    This is all accrual accounting explanation. Converting to market value of assets and liabilities is unnecessary.

    The interesting question is what happens generally to the size of banking spreads and target ROE’s when rates go negative. Positive spreads and ROE will still prevail, but the equity risk premium presumably stretches out as the target rate goes more negative.

  177. K writes:

    JKH,

    Needless to say, I agree. But, why do you envision a widening equity risk premium? Apart from various money illusion effects, I don’t see anything special happening as we pass below zero. Are you saying that market risk premium rises as nominal rates drop, even at positive rates? Why? What if inflation was 10%. Would 8% then be a special rate or would it still be 0%. Equity risk premia might rise if there was signicant inflation uncertainty. But I don’t see that being a function of negative rates.

  178. Dan Kervick writes:

    Challenge accepted. What is the “liquidity operation” that generates equal or superior profit compared to the loan? Just *once* please give me something coherent. Every liquidity operation generates exactly zero profit.

    You haven’t actually accepted the challenge K. If as you say every one of the liquidity operations for the bank generates zero profit, then since the bank’s negative interest loan clearly taken in itself degrades the bank’s balance sheet compared to wherever it was before making the negative interest loan, it seems to me the conclusion is that you have not found an example favorable to your case. It’s not my job to to canvass each and every one of infinitely many possible combinations of operations. It’s your job to show that there is even one case in which (i) the bank engages in some balance sheet transaction X with the Fed; and (ii) the bank engages in some balance sheet transaction Y with a loan customer; (iii) Y is a negative interest loan; and (iv) X+Y > X.

    Note, that is not enough to defend your position to show that there is a case in which X+Y > 0; Because even if X+Y > 0, the bank will instead do X alone if X > X+Y.

    Lets take a virgin bank. It has $1 million of equity and $1 million of reserves. Nothing else. Then it makes a $9 million loan to me. I transfer the money to JP Morgan. What, in you opinion, are the assets and liabilities on the banks balance sheet after these initial transactions are done? What is the accounting value of each of these assets and liabilities (including equity).

    I just want to be clear on what you’re asking K. The bank’s equity is equal to its assets minus it’s liabilities, so when you say the bank is starting with $1 million in equity and $1 million in reserves, you’re are talking about the same $1 million, right, and assuming all of the initial equity has been allocated to reserves?

  179. JW Mason writes:

    K,

    Again, I think you are right in your debate with Dan Kervick.

    All I am saying is that if we order projects by the interest rate at which borrowers find them worth undertaking, the project that become worthwhile when you go from -4% to -5% may look quite different from the projects that become worthwhile when you go from 5% to 4%. In particular, an increasing share of the marginal projects will consist simply of holding various assets as real rates go further below zero. I suspect that in an economy which required an interest rate well below zero to reach full employment, you would in practice see unacceptably large and unpredictable asset price fluctuations before you got there. Arguably this is what happened in the past decade.

  180. K writes:

    jkh,

    I now think I know what you mean about risk premia. Expected real returns are staying fixed (to first approximation) but the real rate is dropping. But this has nothing to do with the zero bound. It’s just the normal effect of monetary stimulus: reduced real interest rates increase expected profits on real investments. That was my argument in comment 158.

  181. JKH writes:

    K,

    If the target rate goes negative, the risk free rate goes negative.

    As I said/inferred, unless spreads compress, and with other things equal, expected ROE increases.

    Doesn’t that mean the equity risk premium increases?

    Maybe those things aren’t equal. But that was the starting point for the idea.

    Interested in what you think about other things not being equal, etc.

  182. JKH writes:

    sorry, meant expected ROE stays the same

    but risk free rate goes negative

    so equity risk premium increases, other things equal

  183. K writes:

    Dan,

    Our disagreement clearly comes down to bank balance sheet operations so I’ll focus on this:

    “when you say the bank is starting with $1 million in equity and $1 million in reserves, you’re are talking about the same $1 million, right, and assuming all of the initial equity has been allocated to reserves?”

    Absolutely correct.

  184. JKH writes:

    On second look, the change in the risk free rate is offset by an equal and opposite change in expected ROE, other things equal (including spread and leverage)

    So the equity risk premium remains the same I guess

    I joined this late – rusty

  185. Morgan Warstler writes:

    Trixie,

    You are exactly the person who’s going to get fixed. And I’m going to fix you.

    Note you need to hide your name and face in order to sound off against the man.

    Once we know who the slackers are amongst us, the $240 is just weekly bait to ID them… we can save all the real workers, anyone who tries and when you get suspended from the dole, you won’t have other unemployed to commiserate with.

    Whether you like it or not, we’re making life very hard for the lawyers these days, you should come see TX. In another ten years, losers will pay.

    Remember Trixie ONCE middle class Americans get a nice sweet taste of making you clean their garage for $3 per hour, they sure as shit are going to vote for laws that keep you from suing them.

    (giggle)

  186. Dan Kervick writes:

    OK, K, here’s the accounting. At the initial stage:

    1. Reserves are $1 million; equity is $1 million; all equity is allocated to reserves. Liabilities are $0; assets are $1 million.

    Now let’s assume the bank makes a $9 million loan to a business. Loans increase liabilities and assets by the same amount. The bank creates a deposit balance for the loan customer which is a liability of the bank; the customer gives the bank a repayment promise which is an asset of the bank. So the result at this stage is:

    2. Reserves are $1 million; deposit liabilities are $9 million; loan assets are $9 million. Total assets are $10 million; total liabilities are $9 million. Equity = assets – liabilities = $1million.

    Note the bank’s capital ratio is solid. The risk weighting of the reserves is zero and the risk weighting of the loan is 1, so the total risk weighted assets are $9 million. The capital ratio is thus 1/9, which exceeds the .8 requirement.

    Now let’s assume, as you say, that the loan customer transfers the entire amount that was loaned to Goldman Sachs. As a result, the customer’s deposit balance goes to zero. A Fedwire transfer generates a $9 million draft on the bank’s reserve account and a $9 credit to Goldman Sachs’s reserve account. As a result, the bank’s reserve account is overdrawn:

    3. Reserves are -$8 million; deposit liabilities are $0; loan assets are $9 million. The reserve balance is now a $8 million liability, the amount the bank owes the Fed. Total assets = $9; total liabilities = $8 million; equity = assets – liabilities = $1 million.

    There is some amount $X that will be imposed as an overdraft penalty until the bank restores its reserve account to the required level. Let’s leave that out for now and add it back in at the end. Assume now the loan customer repays the loan with 5% interest, let’s say by writing a check on its account at Goldman Sachs or some other bank. As a result, $9.45 million is transferred into the bank’s overdrawn reserve account. The result:

    4. Reserves are $1.45 million; deposit liabilities are $0; loan assets are $0. Total assets = $1.45 million; total liabilities = $0; equity = $1.45.

    However, we have to add it whatever the overdraft penalty was. That penalty will be debited from the reserve account:

    5. Reserves are $1.45 million minus $X ; deposit liabilities are $0; loan assets are $0. Total assets = $1.45 million minus $X; total liabilities = $0; equity = $1.45 minus $X.

    Whether the sequence of operations turned out to be profitable for the bank depends on the size of X.

    On the general issue we have been discussing about negative interest lending, I believe you are carrying intuitions about arbitrage into below-zero territory where they don’t any longer apply. For the moment, set completely aside whatever transactions the bank makes in the interbank market and with the Fed. These transactions affect only its reserve account. Focus for the time being only on the negative interest loan transaction itself. That transaction considered only in itself clearly loses the bank money. Now it seems to me that you have been arguing that despite that fact, if the transactions with the reserve account are profitable, then the bank still profits by arbitrage. The bank loses money on the negative interest loan, but it makes money on the reserve operations, and the spread between what was made and what is lost might be positive.

    I have argued in response: Yes, but that still doesn’t mean that making the negative interest loan was rational for the bank. If the reserve account transactions are profitable in total, and the loan transaction is not profitable in total then making the reserve transactions alone will be better for the bank then making the reserve transactions together with the negative interest loan transaction.

    The fact that the spread between transaction X and transaction Y is profitable is no reason to perform transaction X+Y if transaction Y is even more profitable than transaction X+Y.

    If you want to consider only cases in which the reserve account transactions net to zero and are not profitable at all, then it seems to me that you have damaged your case even further.

  187. JW Mason writes:

    Hm. Now I’m feeling less certain about this.

    Is it important that in the real world the $8 million would normally not be an overdraft from the Fed, but a liability to another bank? Or is that detail irrelevant here?

    I’m beginning to think Dan K. might be right here after all. When you frame the question as what other asset will the bank hold, given that reserves pay a negative interest rate and cash isn’t available, it sounds like the bank might rationally choose a negative-yield loan as an asset. But that assumes the size of the bank’s balance sheet is fixed. The alternative to holding a negative-yield asset isn’t holding cash, it’s not-creating a liability, i.e. a smaller balance sheet.

    In that sense, we could say that there is always a zero-yield action available, namely just leaving the existing set of nominal contracts in place.

    So Dan, just to make sure I’m with you, this is an argument that inflation is not neutral, and nominal rates matter? Because I assume we do not want to argue that banks will refuse to lend at positive rates that are less than inflation? i suppose because inflation doesn’t just change the terms of new contracts, it rewrites the terms of old ones, so in that case the “do nothing” option is really not available.

  188. JW Mason writes:

    Yes, I think Dan Kervick is right. A little worrisome that I was so sure he was wrong before. What was the mistake?

    (Maybe it was just thinking that the decision that banks were making was just the composition of the balance sheet, and not also its size? Talking only about composition would make sense in a world where there is only *net* lending so the change in assets is equal to the difference between income and consumption. I suppose that’s the essence of the loanable-funds view — there is a fixed amount of savings that has to be held in one form or another, so people will choose the highest yield even if it’s negative. But as we all know — as I should have been thinking — banks aren’t sitting on a fixed stock of “savings” that they then lend out.)

  189. K writes:

    Dan,

    “There is some amount $X that will be imposed as an overdraft penalty until the bank restores its reserve account to the required level”

    Uh, no. There’s no “overdraft”. If you are short by the end of the day you have to go to the *discount window*. That costs fed funds *plus*. There’s no free money, Dan. Come on!

    JW Mason,

    You’re making a mistake! You’ll be back.

  190. JW Mason writes:

    … altho, on the other hand:

    A “negative interest rate” surely means a negative policy rate, i.e. that the rate at which the Fed makes liquidity available to banks is negative. Regardless of the specific mechanics, if you’re talking about a negative rate, that’s what you should mean. So in Dan’s example at 186, the overdraft penalty should be negative. (In fact it should be equal to the IOR.) This makes the transactions described in 186 more likely to be profitable, and suggests that a negative rate could increase the supply of credit. But I’m not sure….

    Another thought:

    Don’t lots of prices have a kind of ZLB? We just call them abundant. Increase the supply of breathable air all you want, the price never goes negative. I think maybe something similar is going on here, with the interest rate as the price of liquidity.

    Perhaps analogously, think about some activity that can be costly, free, or paid. Say, commenting on this blog. If SRW charged a high fee for the right to comment, he’d get very few comments. If he charged a small fee, he’d get more. With no fee, he’d get even more. So it might seem that if he wanted to increase the number of comments just a little more, a small fee paid to commenters would do it. What’s so special about zero? But of course it’s obvious that (1) the systems needed to implement payments to comments would be very different from the systems needed to implement payments from comments and (2) the type of comments induced by the shift from a “comment fee” of 50 cents to zero, would be entirely different from the comments induced by a shift from 0 to a “comments payment” of 50 cents. In fact the latter would quickly destroy the blog without lots of other changes.

  191. JW Mason writes:

    There’s no “overdraft”. If you are short by the end of the day you have to go to the *discount window*. That costs fed funds *plus*.

    Well, in the normal course of events you don’t go to the discount window either. You borrow your reserves in the interbank market. That’s what the Fed Funds rate *is*. But why does it matter? How does it change the story if we change the word overdraft to discount window or interbank market?

  192. JW Mason writes:

    But let’s consider the discount window case. Suppose the discount window rate is negative.

    Either the the IOR is still positive (or less negative), in which case banks will pile up reserves without limit. Or it is equal, in which case you haven’t increased the incentive for the banking system as a whole to increase credit to the real economy.

    I’m thinking that the best way to think about this, is to remember that the interest rate is the price of liquidity. Bank liquidity is normally a constraint on lending. Monetary policy operates by changing the supply of liquidity, targeting its price (the interest rate). But there is no way to make liquidity (or anything else) so abundant that its price goes negative. If you try, you either end up with an infinite quantity demanded, or it ceases to be a binding constraint. Presumably in a liquidity-abundant world the main constraint on lending is bank capital.

  193. K writes:

    JW Mason,

    “In that sense, we could say that there is always a zero-yield action available, namely just leaving the existing set of nominal contracts in place.”

    OK. Borrowing money in the interbank market at -5% and lending it commercially at -4% is not zero yield. It’s profitable! So why would they leave their balance sheet doing nothing. Come on, guys! It’s not about the negative rate. It’s about the spread between borrowing and lending. Anyways, you seem to still be debating this with yourself!
    :-)

    “How does it change the story if we change the word overdraft to discount window or interbank market?”

    I don’t care. But Dan tried to leave as a mystery charge “X”, cause otherwise he’ll be wrong. It’s an interest rate, fed funds or discount rate.

  194. Dan Kervick writes:

    I didn’t say there was free money K. The overdraft isn’t free. It’s a liability of the bank – money it owes to the Fed. And it only comes with the penalty rate $X. The bank pays that money back when its collects on the commercial loan.

    It doesn’t matter how you parse it: If you want, divide the bank’s account at the Fed into its reserve account and its discount window account. To make the payment to Goldman, the bank borrows $8 million from the Fed at the rate of interest $X. The discount window loan is secured with the $9 million commercial loan as collateral. $8 million is credited to the reserve account and $8 million + $X is debited from the discount window account. The $8 million in the reserve account is added to the $1 million that is already there. The $9 million payment is made to Goldman bringing the reserve account to zero and leaving the bank $8 million + $X in the hole with its discount window account. Then when the bank collects the $9.45 million when the loan customer pays off the loan, part of that sum is used to clear the negative balance discount window account, and the rest is deposited in the reserve account.

    I just combined all of this into one account, which is the way I understand it actually works: automatic overdrafts plus a penalty rate. But if I am wrong about that and there are two accounts involved, it doesn’t matter to the overall balance sheet with the Fed.

  195. Dan Kervick writes:

    Because I assume we do not want to argue that banks will refuse to lend at positive rates that are less than inflation?

    Yes, JW Mason, I believe you are right. The nominal rate does matter. If I have 9 units of anything right now and my choices are between having 9 units of that thing one year from now and 10 units of that thing one year from now, then it doesn’t matter whether the unit value of the thing in question rises or falls between now and a year from now, I will always prefer having 10 units of it to 9 units of it. (I’m assuming that we know the thing in question will always have some non-negative value, even if the value falls – it’s not nuclear waste, for example.)

    And if I already have 10 units of that thing right now and my choices are between having 9 units of that thing one year from now and 10 units of that thing one year from now, then again it doesn’t matter whether the unit value of the thing in question rises or falls between now and a year from now, I will prefer having 10 units of it to 9 units of it.

  196. Dan Kervick writes:

    I don’t care. But Dan tried to leave as a mystery charge “X”, cause otherwise he’ll be wrong. It’s an interest rate, fed funds or discount rate.

    I just left that number blank because we hadn’t specified any actual discount borrowing rate and penalty rate. But in the example as given, so long as the total $X turns out to be less than $450,000, the bank ends up with a better balance sheet at the end than at the beginning.

    Right now, I believe the primary credit discount rate is .75%. Thus the rate for borrowing $8 million would be only $60,000.

  197. JKH writes:

    Discount window borrowing “on its own” cannot be “profitable” in the normal course, whether rates or positive or negative. That’s because the discount rate is always higher than the fed target rate and/or IROR, whether rates are positive or negative.

    The marginal discount window transaction is borrowing reserves from the window, thereby creating reserves as an asset. The discount rate is higher than the fed target rate/IOR, so the spread is negative by definition. E.g. the spread is negative, whether the two rates are 1 and 1.75, or are (1) and (.25). The spread is (.75) in both cases. So discount window borrowing is always unprofitable at the margin.

    It works the same way for loans already on the books, where funding previously in place is lost. Loans already on the books are priced at an economic spread to cost of funds – i.e. at a spread to the target fed rate in the case of a floating rate loan. Discount window borrowing indicates a loss of market rate funding – i.e. funding at the target rate. Using the negative rate example, discount window borrowing is a net cost at the margin, because the spread between the loan rate and the discount rate of (.25) is always less than the spread between the loan rate and the fed target of (1). The spread has temporarily compressed due to discount window borrowing.

    So discount window is borrowing is a net cost whether it produces net new reserves or replaces reserves lost when previous market funding is lost.

    The probability of required discount rate borrowing is typically considered as negligible in loan pricing at the outset, so it has negligible effect on original spread pricing. Loan pricing assumes normal market borrowing and spreads.

    (Fixed rate loan analysis is entirely different, since such loans are normally hedged either with fixed rate funding or interest rate swaps.)

  198. Trixie writes:

    Grasp those straws, Morgan. Grasp ‘em! Grasp ‘em!

    Somehow I’m still not a-scared. See you on the front line.

    #PickleJars

  199. Dan Kervick writes:

    One thing I think we should also remember is that our “virgin bank” thought experiment here is, while an entertaining simplification to think about, completely unrealistic in understanding a bank’s approach to its loan decisions. We are imagining that the bank has all of its deposit liabilities in a single account that is 9/10ths the size of its total assets, and that the possibility therefore exists of a transfer of 100% of its deposits to another bank in a single day, leading to a massive overdraft on its reserve account. Obviously this is not the way things work in the real world, which is why the required reserve ratios is 10%, and not 90% or 100%.

  200. Morgan Warstler writes:

    Trixie, they used to say never pick a fight with people who buy ink in 55 gallon drums.

    Now, the same thing should be said about picking fights with folks who have 200M guns. That’s why I get so frustrated with progressives and their end of times junk.

    Steve’s whole screed is just barely scratching the surface of how strong the American foundations are. Folks who actually have already delivered the goods vs. those who just insist they will is never a Davis and Goliath story, because once the noobs get done running the gauntlet they have changed.

    They changed the EXACT same way my plan will change others. People improve their lot, they look back and are humiliated by the folly of their youth.

    How in the hell do you think the baby boomers lean so far right? The hippies all watch Fox News. They din’t sell out, they wised up.

    But I like you, I will bid you up, and leave you great feedback. You will be my jester. I like art.

  201. JW Mason writes:

    Borrowing money in the interbank market at -5% and lending it commercially at -4% is not zero yield. It’s profitable!

    But this is not a description of any actual transaction. We have to be more precise in our language — “money” is not a meaningful term here.

    What the bank borrows at the discount window is *reserves*. It increases the reserves on the asset side of its balance sheet and adds a debt to the Fed on the liability side. It does *not* lend these reserves to nonbank borrowers — as you know perfectly well, reserves never leave the banking system. There is a second, logically independent, transaction in which the bank creates a deposit and a loan. You have to explain why undertaking the first transaction makes the second one more attractive.

    Now, you could say, the rate on the liabilities the bank acquires in the course of making a loan will follow the rate on discount window borrowing. But that is not plausible, because the liabilities the bank creates in the course of lending are liabilities to the nonbank public. And why should the public hold bank liabilities paying a negative return when they can get close to a zero return simply by holding a stock of any asset with low carrying costs?

    From the point of view of the credit supply to the nonbank economy, what matters is the *difference* between the cost to banks of acquiring reserves and the return from holding them. As JKH says, usually that is positive (the second term is normally zero) so banks try to minimize their reserve holdings. This leads them not to make loans they would otherwise make, which is how monetary policy is able to affect the supply of credit to the real economy. The difference cannot be negative, since banks would then demand an infinite quantity of reserves. The best you can do is the current situation,w here banks are indifferent as to the size of their reserve holdings, since the cost of acquiring them and the return on them are identical. in that situation, reserves simply cease to be a binding constraint on loans. Talking in terms of a homogeneous stock of “money” that the Fed lends to private banks, and private banks then lend to the rest of us, just confuses things.

    And yes, I am still arguing with myself. :-) It may make me look a little foolish, but it also makes me hopeful that I am learning something.

  202. JW Mason writes:

    Let’s think about this some more.

    Suppose tomorrow the Fed instituted a 5% tax on reserves, and set the discount window rate to -5% (or -4.5% if you like.) Finally, they announce that they will no longer exchange reserves for cash at par. I think we all agree this would constitute instituting a negative policy rate?

    This implies a capital loss for banks holding large quantities of reserves. They will now wish that they had made more loans in the past, which would have shifted those reserves to other banks. But does it change the incentive to lend going forward? I don’t see why it should.

    It’s true that banks with excess reserves will now be willing to lend them in the interbank market at negative rates down to -5%. So on first glance that should make funding costs lower for new loans. But the problem is, the borrowing bank will now incur the tax, bringing its cost of funding back up to zero. So the effective interbank interest rate is zero, not -5%.

    The fundamental problem here is that the supply of reserves is infinitely elastic at the price set by the central bank. So for equilibrium with a finite quantity of reserves held by the banking system, the price of reserves (i.e. the policy rate) cannot be lower than zero.

  203. JW Mason writes:

    Altho, sorry … there is another point of view. If a bank could borrow reserves at a negative rate and then induce another bank to hold those reserves, obviously that would be profitable. So the transactions would look like (1) Bank A acquires $1 million reserves (paying a -5% interest rate) and a $1 million liability to the Fed (also at -5%). (2) Next, it makes a loan of $1 million at -2% to a private borrower, creating a deposit paying 0%. Now Dan Kervick here will say that this second transaction is obviously money-losing, why would nay bank do it? But what if (3) the borrower then transfers the deposit to Bank B, so both the deposit (liability) and the $1 million in reserves (asset) move to bank B’s balance sheet.

    So at the end of the year, Bank A has a profit of $20,000 andBank B has a loss of $50,000. (Negative rates are bad for banks.) But Bank A seems to have been following incentives in creating new loans, as K has been arguing. So does this story make sense? I’m genuinely not sure.

    First of all, what compels Bank B to accept the deposit ad the reserves? Maybe regulation … but then won’t this policy accelerate the replacement of regulated banks with unregulated quasi-banks? Second, more broadly, the profits to A here come not from making a loan per se, but by *replacing its deposits with liabilities to the Fed.* That seems like the direction a negative-rate policy, if it could be effective, would lead: Not to negative lending rates to nonbank borrowers, but to the elimination of bank liabilities to anyone but the Fed, or in other words, the exit of banks from the intermediation business.

    Now, you might say that the liquidity/payment services provided by banks are so valuable that if banks threatened to stop taking deposits, depositors would be willing to accept negative rates to continue having access to those services. But given all the proliferation of nonbank financial institutions, it seems much more likely that deposits would just migrate to nonbank institutions without access to the Fed window, which would still be willing to pay nonegative rates for them.

  204. JW Mason writes:

    (Ooops, A should have a profit of $30,000.)

  205. K writes:

    JW Mason,

    “But this is not a description of any actual transaction.”

    It kind of is. In detail though…

    Bank A has $1 million reserves and one million of equity. They lend me $9m creating a $9m loan asset and a $9m deposit liability. I transfer the deposit to JPM, meaning bank A sends a $9m of reserves via Fedwire to JPM and borrows $9m via fed funds (ultimately from JPM but doesn’t matter). Now Bank A has 9m loan and 1m reserve assets and 9m interbank loan and 1m equity liability. That’s the way it typically works. Now they could also have gotten the missing funding via the discount window. Whatever. The point is, and this is at the root of my dispute with Dan, whether the liabilities are discount window, interbank, or deposit, the new 9m loan asset will be accompanied by 9m of new liabilities. Those all pay about -5% in our example (deposits slightly less, discount window slightly more). I don’t know what you don’t get, Dan. It doesn’t matter that some items are reserves and others are deposits. What matters is that assets=liabilities and commercial loans earn more than deposits or interbank liabilities (which in turn yield more than reserves making borrowing to hold reserves totally unprofitable, as JKH points out). From the bank’s perspective it really is that simple. Consumer lending is profitable, reserves are not. The sign of the policy rate is *utterly* irrelevant. I don’t know what else to say.

  206. K writes:

    JW Mason,

    “But does it change the incentive to lend going forward?”

    No! Absolutely not. That’s not the point. Nothing changes for the bank. But let’s say *I* have a project costs 100 consumer baskets to begin, and which I expect to pay off 102 consumer baskets in one year, then if the banks drop my real borrowing rate to somewhere well below 2% that makes my project more enticing to me and I might borrow money to do it. Stimulus works via incentives on consumers/investors. Not via incentives for banks.

  207. K writes:

    JW Mason,

    I laid out the detailed stimulus example in 158.

  208. JW Mason writes:

    Those all pay about -5% in our example

    But I’m not convinced that is true.

    Either the discount rate is less than (i.e. more negative) than the interest on reserves, in which case banks will hold an infinite quantity of reserves. Or else the discount rate is equal to the IOR, in which case the effective rate faced by the borrower is zero.

    i don’t think you can actually get a negative rate in the interbank market. The Fed achieves its target rate by standing ready to supply whatever quantity of reserves banks demand at that rate. If the target rate were negative, the quantity of reserves demanded would be infinite.

  209. JW Mason writes:

    if the banks drop my real borrowing rate to somewhere well below 2% that makes my project more enticing to me and I might borrow money to do it. Stimulus works via incentives on consumers/investors. Not via incentives for banks.

    Sure, if you could exogenously fix the rate set by banks. But you can’t, as long as we have a private banking system. So that’s where incentives for banks come in — determining where they set your borrowing rate.

  210. Dan Kervick writes:

    The point is, and this is at the root of my dispute with Dan, whether the liabilities are discount window, interbank, or deposit, the new 9m loan asset will be accompanied by 9m of new liabilities. Those all pay about -5% in our example (deposits slightly less, discount window slightly more). I don’t know what you don’t get, Dan. It doesn’t matter that some items are reserves and others are deposits.

    I lost you here K. What is it that pays -5%? Isn’t -5% the interest on reserves? Reserves are the bank’s asset, deposits are the bank’s liabilities. The bank doesn’t earn interest on its liabilities. If, I’m a bank customer, and I have $1,000 in my deposit account, then that balance is my asset and my bank’s liability. The Fed doesn’t pay interest on balances in that account. On the other hand, if I’m the bank, and I have $1,000,000 in my reserve account at the Fed, then that $1,000,000 is my asset and the Fed’s liability. The Fed does pay interest on that account, in a system with interest on reserves.

    If Ebeneezer Scrooge has a ledger that says he owe’s Bob Cratchit 5 Pounds with interest, and Bob Cratchit has a ledger that says he owes Tiny Tim 2 Pounds with interest, that doesn’t mean that Scrooge owes Tiny Tim interest on his 2 Pounds.

    If a commercial loan to a business earns more interest for the commercial bank than the Fed earns from a loan to the commercial bank, then the bank can profit by borrowing from the Fed and lending to the business. And if, in negative interest world, the commercial bank loses less interest by lending to the business than the Fed loses by lending to the commercial bank, then the commercial bank can again profit by borrowing from the Fed and lending to the business.

    But in that latter, negative interest world, the commercial bank could earn even more profit by borrowing from the Fed and not making the loan to the business.

  211. search engines…

    [...]interfluidity » Depression is a choice[...]…

  212. JW Mason writes:

    In that latter, negative interest world, the commercial bank could earn even more profit by borrowing from the Fed and not making the loan to the business.

    Dan,

    That you kept saying this is one of the reasons your argument seemed wrong to me initially. This statement only makes sense as an argument for why negative interest rates cannot exist at all. If a bank can hold reserves at zero cost, and borrow from the Fed at a negative interest rate, it would hold an infinite quantity of reserves. So we have to assume that the IOR is at least as negative as the rate the bank can borrow at.

  213. K writes:

    “So that’s where incentives for banks come in — determining where they set your borrowing rate.”

    No. They set the *spread* to the risk free rate. The cb sets the risk free rate. Yes???

  214. JKH writes:

    REPEAT:

    “The discount rate is always higher than the fed target rate and/or IROR, whether rates are positive or negative.”

    PLUS – IROR is always lower than or equal to the fed target rate.

    Those two things are essential to the Fed controlling the target rate trading range.

    PLUS – banks lend as a spread over the fed target rate (or a fixed rate equivalent)

    PLUS – banks don’t normally price that spread assuming borrowing at the discount rate

    PLUS – all that works at positive or negative rates

    =

    Banks can’t profit in the normal course from borrowing at the discount rate and “doing nothing”

    It’s always a marginal cost based on the structure of rates

    That works at positive or negative rates

  215. Peter K. writes:

    Dan Kervick insists that inflation is a regressive tax on wages.

    Dean Baker addresses the isssue in a blog post:

    http://www.cepr.net/index.php/blogs/beat-the-press/how-is-inflation-a-highly-regressive-tax-on-wages

    Make sure you look at the addendum which has a chart of compensation and inflation. High unemployment is a regressive tax on wages.

  216. Dan Kervick writes:

    If a bank can hold reserves at zero cost, and borrow from the Fed at a negative interest rate, it would hold an infinite quantity of reserves.

    Yes, I agree. To make such a system work, the CB would have to set some kind of cap on borrowing reserves.

    I think I also mentioned somewhere back at the beginning of the thread that one way to make this whole negative interest business work might be to make reserve borrowing conditional on commercial loans. Suppose every time a bank grants a loan application, the bank and the customer co-sign some special kind of Fed voucher with a face value of 10% of the loan. The bank can then take that voucher to the interbank market or the discount window and borrow the amount of reserves on the voucher. In that case, the bank can’t profit from its transactions with the Fed unless it makes loans to customers. Competitive pressures might then push commercial interest down below zero. But as long as the reserve transactions and the loan transactions are independent, the bank profits more by doing the reserve transaction without the loan transaction.

    Of course, if the Fed really wants commercial borrowers to get sub-zero interest rates, it could lend them the money itself, and cut the banks out of the action. The zero nominal bound seems grounded in the fact that the banks are profit-making businesses, not philanthropic institutions. But the Fed can follow its full employment mandate and play philanthropist if it wants.

  217. Dan Kervick writes:

    Banks can’t profit in the normal course from borrowing at the discount rate and “doing nothing”

    OK JKH, but if that is the case, then it also can’t profit by borrowing at the discount rate and making a negative interest commercial loan, since lending at negative interest is by definition less profitable than doing nothing.

  218. K writes:

    Dan

    All I’m saying is that it doesn’t matter that you can’t lend out reserves. Ie. reserves can’t be directly turned into commercial deposits. All that matters (which is what you disputed for a long time) is that assets have to be equal to liabilities, and commercial loan asset interest is higher than the interest on their liabilities (deposits, ff or discount window). So commercial lending is profitable.

  219. K writes:

    Dan,

    “since lending at negative interest is by definition less profitable than doing nothing.”

    Dan!!! For the millionth time, *NOTHING IS NOT AN OPTION*. There must be an asset! Assets=Liabilities! I’m dying here. That’s why I made you do that stupid example. JKH please save me.

  220. JW Mason writes:

    Dan K.,

    Right. I think we agree now.

    JKH,

    So how does the Fed control its target rate? I would have said, by standing ready to buy or sell any quantity of reserves at that rate.

    And what quantity of reserves would the Fed have to sell to achieve a negative rate? I would have said, more than an infinite quantity.

    I take it you disagree, but why?

  221. JW Mason writes:

    K,

    Explain this to me. If the IOR is at least as negative as the discount/interbank rate, isn’t the effective cost of funds to the borrower nonnegative? And if the IOR is 0, or at least less negative than the discount rate, wouldn’t banks choose to hold an infinite amount of reserves?

    More broadly, how does the Fed set a price in the interbank market (how does one set a price in any market) except by adjusting quantities? And what quantity of reserves would give them a negative price?

  222. Dan Kervick writes:

    OK K, I think you’ll agree we’re going in circles now, and we’ve probably reached the point where its unlikely we’re going to convince each other of anything soon. Thanks for all the time you’ve taken discussing this issue. I’m going to try to boil down my argument into a blog post for New Economic Perspectives, and then maybe we can continue the discussion there.

  223. K writes:

    JW Mason,

    The price we are discussing is the price of $1 in a year. It usually costs less than a dollar, but it could cost more. It obviously can’t be negative. The interest rate is the log of the price divided by the tenor. It could be positive or negative. Whatever.

    “And if the IOR is 0, or at least less negative than the discount rate, wouldn’t banks choose to hold an infinite amount of reserves?”

    IOR <= Fed funds <= Discount rate. Always.

  224. K writes:

    Dan,

    “maybe we can continue the discussion there.”

    You bet. See you next time!

    K

  225. girls of desire…

    [...]interfluidity » Depression is a choice[...]…

  226. K writes:

    FW Mason,

    No matter what the policy rate, banks will prefer to hold zero reserves, so long as IOR <= Fed funds <= Discount rate. (They have a choice between keeping excess reserves and lending but they can't do "nothing"). If lending earns more than Fed Funds, they'll lend. The sign of the policy rate is irrelevant.

  227. [...] interfluidity » Depression is a choice [...]

  228. JKH writes:

    “I’m going to try to boil down my argument into a blog post for New Economic Perspectives”

    friendly suggestion – run it by Fullwiler first; your argument here is not correct, but maybe you can modify it in a post somehow

  229. K writes:

    JW Mason,

    “what quantity of reserves would give them a negative price?”

    Any quantity is possible, but only if they let fed funds drop to the level of IOR. Otherwise banks will all try to dump their excess reserves at the ff rate, and ff will drop to IOR. Other than that, banks will hold zero excess reserves. Holding reserves that earn less than the fed funds rate is a losing trade right? The fed achieves the target by the the fact that if banks trade off-target, the Fed can force the rate back on target by the close, thus ensuring that one of the parties to the off target trade loses money. Everyone knows they can do this, so nobody offers funds below target or bids them above. The CB rarely, therefore, actually has to follow through on it’s threat, so excess reserves tend to stay at zero. Nothing special happens at or below zero. Excess reserves are only occurring right now because the fed wants them so they have floored the fed funds rate at the effective IOR rate.

  230. K writes:

    JKH,

    “run it by Fullwiler first”.

    Oh, thanks a lot, man. Now there’s not a chance we’ll ever see that post.

    Dan, when that post never shows, I’m going to take it as acknowledgement that I was right. It’s all I’ll ever get, so I guess I’ll have to take it.

  231. [...] mistaken theory, institutional frictions, personal quirks, and political forces rather than being, as I argue, a choice. I’d be more sympathetic if these “mistakes” were unique to the United [...]

  232. JKH writes:

    Maybe I shouldn’t have suggested that.

    Run it whatever way, if you’d like to see more free discussion and exploration of the point.

  233. [...] http://www.interfluidity.com/v2/3212.html FacebookTwitterMoreDiggStumbleUponRedditEmailLike this:LikeBe the first to like this post. This entry was posted in Uncategorized. Bookmark the permalink. ← Sunset Photo [...]

  234. JW Mason writes:

    The price we are discussing is the price of $1 in a year.

    That’s one way of looking at it. Another way is Keynes’s way: The interest rate is the price of liquidity.

    Note that in practice the Fed does not operate on the margin of a dollar today for a dollar tomorrow. That is a much larger market, which is why QE has to be so huge (and still is not very effective. The Fed normally sets the price of reserves, which are important (a) because they are liquid and (b) because of their special role both by regulation and as the means of interbank payments.

    They have a choice between keeping excess reserves and lending but they can’t do “nothing”

    No, that isn’t right, except insofar as increasing loan volume turns some of the excess reserves to required. (But this can happen without making any new loans, too.) The volume of loans made by banks, and the volume of reserves held by banks, are in principle entirely independent. The link comes only to the extent that reserve requirements (statutory or prudential) bind, which at the moment they do not.

    The aggregate volume of reserves held by the banking system is completely unaffected by any loan decisions.

    Do you agree that if there were only a single bank, then instituting a negative policy rate, IOR and discount window rate (with discount >= fed funds >= IOR if you prefer) would have NO effect on the interest rate or volume of lending to nonbank borrowers?

    If you don’t agree, then we can discuss the case of a single bak, which may be simpler. If you do agree, we can explore what changes when there are multiple banks.

    Excess reserves are only occurring right now because the fed wants them so they have floored the fed funds rate at the effective IOR rate.

    I don’t agree with this. There are excess reserves now because the Fed had to create new reserves to fund its purchases of bad assets in the various bailouts of 2008-2009, since the volume of assets it was buying from the banks threatened to exhaust its Treasury holdings.

    Changing the policy rate would not affect the volume of excess reserves. How could it?

    I still think you are confusing reserves, which are used only for interbank transactions, with the “money” that is lent to nonfinancial borrowers. The Fed’s assets and liabilities have to balance, just like everyone else’s. So as long as all those MBSs and whatnot are on the Fed’s balance sheets, the volume of reserves held by the banking system cannot fall.

  235. Sergei writes:

    Guys, put it up on NEP regardless of Scott! I had no energy to go through your ping-pong and it is no wonder there can be confusion. Negative interest rates is not something anybody can have a clear intuition into out of the box. Like JW opinions can easily changes sides. Dan, do not check with anybody but your common sense! Just put it there for everybody to read! If needed Scott will log in to post the truth.

  236. Sid writes:

    I don’t think the median voter is necessarily the elderly. Intuitively this definition seems to be more fluid, perhaps a function of {largest age cohort, wealth, organisational ability/lobbying}

    Emphasising the median voter also suggests that policy making in a western democracy is merely a clearing house for interests. When in reality it isn’t completely like that, what with corporates, ideologues and other interest groups.

  237. Stefan Stackhouse writes:

    Some of us are at least somewhat aware of world history, and know that pensioners have been left starving by governments that either default on their obligations, or hyperinflate them away, or both. It has happened within recent memory – see the recent history of the former Soviet Union for just one example.

    Thus, the oldsters and those of us who are rapidly approaching that state are terrified of inflation. Why? Well, if you think that just going back to work is an option, think again. Employers don’t even want to keep people in their 50s on their payroll, why should they want to hire someone in their 70s or 80s? Assuming that the oldster can even get to work and put in a full day, which is doubtful. One’s body parts do start getting serious problems as one ages, you know.

    So my question: What would you have us old and soon-to-be-old folks do? And what should we have done these many years past?

    It is not a purely abstract question for you, because you will most assuredly be in the same shoes soon enough (unless something unfortunate kills you off young).

  238. Becky Hargrove writes:

    Stefan,
    We can recreate knowledge and skills wealth measured by direct time in individual to community aggregates, instead of as a residual in monetary production. That way, no inflation is necessary. No one is speaking of encouraging inflation but rather of maintaining monetary flow for the already existing efforts of those participating in the monetary marketplace. As to the potential for knowledge and skills wealth, there would be a new place in the economy for any who find themselves declared ZMP in the traditional monetary sense.

  239. Tao Jonesing writes:

    “We, collectively, are making a choice.

    In theory, yes. In practice, no. We are not allowed to determine what the choices are for ourselves. We are simply allowed to choose between choices that are provided to us, and those choices all lead to the same general outcome. At best, the polity exists to ratify the policy, not to set it.

  240. [...] his Volcker moment. He has had many opportunities and whether because of groupthink at the Fed, political power of savers, or a failure by him to read Scott Sumner’s blog, Bernanke cannot seem to find his [...]

  241. Rodney Hytonen writes:

    The only wise comment I see here is by Stefan Stackhouse, http://www.interfluidity.com/v2/3212.html#comment-24846

    Remember that due to the mentioned creping ageism (now I believe unemployability may have dipped into the upper 40′s) and the 1-2 year periods of unemployment since the early 1980′s,
    almost NONE of the elderly have any more than sub-FPL Social Security to live on.

    And the so “pre-endowed” are becoming more rare in an increasing population daily.
    Whoever, be it Ryan, SImpson, or Bowles, chooses to CUT anything from the (ALSO btw “Pre-endowed” -by US-) ;entitlments,’ need also to provide funding to clean up ALL those dead bodies frozen and/or starved in their homes next winter.

  242. [...] via interfluidity » Depression is a choice. [...]

  243. Alex writes:

    This article is based on faulty premises:

    1) That monetary or fiscal policy to create more inflation magically creates more economic growth. (This is untrue. Money printing does not create any economic value. And a deeply indebted government borrowing money from the future actually impedes economic growth.

    2) That our policymakers have been too timid with monetary and fiscal policy.
    (In fact, they have been more aggressive than at any time in history. The Federal Reserve has not only cut rates to zero, but htye have done all kinds of unprecedented things. The federal government is spending roughly TWICE as much as it is bringing in in taxes every year–borrowing and spending at an unprecedented rate.)

    3) That the median influencer is not a billionaire (top .01%) but rather just an affluent older person (say, top 5%).
    (This is not true. Affluent savers are being gutted in order to protect the top .01%.)

    You are right that the polity has scrambled to protect the big creditors (banks, their bondholders, and their executive teams–all part of the top .01%). And you are right that this hurts you (someone who hasn’t endowed his future consumption). The government has essentially mortgaged our future by throwing money into the black hole of insolvent banks. And you are also right that the baby boomers will get the better of young people, by trying to suck down as many benefits as they can even if society can’t afford them.

    But you are wrong about the solutions needed. Inflation will help to save the bankers and the very rich (because they get the counterfeit dollars *first* and therefore cn spend them before people realize they are worth less). Inflation also helps the government itself, because it is a stealth tax on savings and on productive workers. It will hurt people who rely on government benefits, because the cost of living will go up by more than the benefits. The government wants inflation, the bankers want inflation, and indebted homeowners want inflation. Young people, productive people, etc.–we’re the ones who get hurt. The bankers and the government have been trying as hard as they can to increase inflation–it just hasn’t kicked in fully because of the deflationary impact of a collapsing credit bubble. As young people trying to get ahead, we should have wanted deflation–then we could buy things we need like houses much cheaper.

    You’re absolutely right that the richest and the government are not acting in our best interests, but your notion that it’s because they are being too timid monetarily and fiscally doesn’t make sense. They are being more aggressive than they ever have, and it’s not for your benefit. Inflation can’t spark economic growth. Money printing is not value creating.

  244. Big Ern writes:

    Fine – if you want to deprive those who hold wealth of that wealth (for whatever reason), why not do it in out in the open in broad daylight with taxation, instead of obscuring the confiscation through inflation? Surely if your justification is noble, it shouldn’t be necessary to redistribute the wealth in secret.

    It would also be better for the credibility of those who guard the sanctity of the currency.

    It would also be surgical, since those who have wealth, but only a modest amount (like me) needn’t have any wealth confiscated at all in a targeted taxation scenario.

    For these reasons, I am totally against what you are surmising, or arguably proposing.

  245. [...] This is perhaps the best explanation of the core reasons of the economic travails of the United States, Japan and many other “developed” economies. The ailing developed economies are plutocratic democracies. “The people” do have power, but influence is weighted in a manner correlated with wealth. The median influencer in these economies is not a billionaire, but an older citizen of some affluence who has mostly endowed her own future consumption. She would like to be richer, of course. But she is content with her present wealth, and is panicked by the prospect of becoming poorer. For such a person, the depression status quo is unfortunate but tolerable. The risks associated with expansionary policy, on the other hand, are absolutely terrifying. [...]

  246. I intended to create you one little observation to finally give thanks the moment again for all the amazing thoughts you have featured in this case. This is so shockingly generous of people like you to supply publicly all that many people might have advertised as an electronic book to help with making some profit for themselves, most importantly considering the fact that you could possibly have tried it if you ever decided. These suggestions also worked to provide a fantastic way to recognize that many people have the identical desire the same as mine to know the truth good deal more when it comes to this matter. I am certain there are several more enjoyable times in the future for individuals who looked over your blog.

  247. Rabbit Bowls writes:

    Rabbit Bowls…

    [...]interfluidity » Depression is a choice[...]…