Financial system failure and the paradox of thrift

Over the weekend, Paul McCulley offered a thought-provoking piece (ht Justin Fox, Brad DeLong), which starts with a discussion of the “paradox of thrift”:

For those of you who might not recall, the paradox of thrift posits that if we all individually cut our spending in an attempt to increase individual savings, then our collective savings will paradoxically fall because one person’s spending is another’s income – the fountain from which savings flow.

There’s a hidden assumption in the “paradox of thrift” that really ought to get teased out more often. It is true that one person’s spending is another person’s income. But it does not follow that an increase in saving translates to a decrease in aggregate income. There are two kinds of spending, consumption and investment. Laying a subway line adds to somebody’s income as surely as buying a Ferrari does. Ordinarily, nearly all savings are actually spent on investment goods, and there is no “paradox of thrift”. What is “saved” is really spent on current production of future capacity, and there are plenty of paychecks to go around. There is no “fallacy of composition“: individually and in aggregate, today’s thrift lays the groundwork for tomorrow’s abundant consumption.

However, for this to work out, two things must be true: Today’s savings must be invested in projects that will actually generate future wealth, and savers must believe they will retain a stake in the increased wealth commensurate with the size and wisdom of their investments. We have a financial system in order to make these facts true. If the investment industry is capable of finding or initiating projects likely to satisfy future wants, and if financial claims are predictable and stable stores of value, we need not trouble ourselves over the paradox of thrift. The issue only arises when the financial system breaks down. When investors lose faith in the quality of available investments or their ability to collect the proceeds (in real terms), they pull out savers’ Plan B: precautionary storage. They buy gold, or oil, or art, or whatever, and they keep it, generating scarcity rents for those who can offer perceived value stores, but very little in the way of general income and employment. Precautionary storage, not thrift itself, is the villain of the tale.

The vulgar Keynesian prescription is to encourage consumption, when a dynamic of precautionary storage takes hold. And in extremis that might be a good idea, because if all everyone does is hoard, it’s hard to figure what to invest in, except maybe storage tanks. But it’s much better to develop a financial system that actually performs, that identifies fruitful projects and allocates claims fairly. Storage eats wealth, while productive enterprise creates it. People know this. No one “invests” in gold or oil when a financial system is working. They do so when it is broken. Like now.

Encouraging people to go shopping in order to help the economy is not “second best” policy. It’s a desperate last resort. We’re not at a point where there’s so little economic activity that we can’t foresee future wants. We’re at a point where people are beginning to shift from investment to storage because of a well-deserved loss of confidence in the financial system. Encouraging consumption now is nihilistic. It feeds into a vibe (I feel it personally, do you?) that saving is so uncertain and money so volatile that one might as well spend, ‘cuz who knows what tomorrow might bring. The right way to sustain aggregate demand and maintain current income is to figure out what we should be investing in — not stocks, bonds, or CDOs, but factories, windmills, or schools — and then to put current resources to work. Our financial system is failing spectacularly because it erred grievously. It built homes and roads and sewers that oughtn’t have been built, it “invested” in vacations and plasma televisions, and it paid itself handsomely for doing so. That’s not a problem we can spend our way out of. To fix the financial system we have to change it, not rally to its support. We will know we’ve put things right when thrift is something we can celebrate, when we save because we are excited about what we are creating rather than frightened by what we might lose.


60 Responses to “Financial system failure and the paradox of thrift”

  1. Alessandro writes:

    Very clear and insightful as always! I’ll point to this article when people ask me ‘What’s happening exactly?’

  2. Alessandro writes:

    BTW: am I right that if the profound problem is the attitude of the savers, all FED and Treasury actions are just plain useless? Alphabet soup facilities, bailout plans, private debt transformed into public debt (aka stimulus checks)? Not one signle action went in the right direction.

  3. Alex R writes:

    This is an elegant essay, that clarifies for me much of the confusion I have about the social value of spending versus saving. If we look at the available investment options as a whole, we see capital investments — railroads, factories, schools — as both continuing current incomes and increasing future incomes. But much of today’s investment goes to lend money to people for current consumption — car loans and credit cards — and for home loans. A home loan has aspects of a current consumption loan, and also has aspects of investment in “storage”. It doesn’t seem to me, though, too much like a capital investment. If loans to buy new houses allows more *needed* housing to be built, that will increase future incomes somewhat like capital investment, but the emphasis must be on the *needed* and the price being paid.

    So, if I put my money in my local thrift, which lends it out to home buyers, is my money truly *invested* in the economy in a way which creates value?

  4. JKH writes:

    The more I read Paul McCulley, the more respect I have for him. The core of his essay on the paradox of deleveraging was marvellous.

    With regard to the paradox of thrift, I think the fallacy of composition holds at the first derivative level. Attempts to change the level of saving too quickly in aggregate will result in new investment in the form of inventory accumulation, reflecting excess production relative to expenditure. This excess production and inventory accumulation may include both consumer and investment goods. In any event, too rapid a change in aggregate savings will start a contractionary process that will lead to reduced aggregate output and income. Once the market clears, the economy will produce a mix of new consumer and investment goods at an “equilibrium” level that reflects the equivalence of investment and saving, (true at the global level; and true at the national level, ex external imbalances). But the overall level of GDP will be lower.

    In terms of the larger question, it really isn’t possible to assess the appropriateness of the consumption/investment mix without first and foremost considering it at the global level. There are no closed economies so there are no closed answers. The US doesn’t save enough to finance the investment it has. One must address this issue before getting to the quality of the investments it has made.

    And the financial system isn’t the system that is making the investments. The financial system is a system of “inside assets”. Economic investment is an “outside asset”. The financial system is about intermediation of savings and investment – not investment per se. And it is a system of multiple and compound and in some cases excessive intermediation.

    The problem of the credit crisis is being viewed indirectly as one of too much leverage. It is better viewed head on as one of inadequate capital – for both financial institutions and households (down payment adequacy). There are many actors to blame, but the one that comes to front of mind in this overarching context is the one logically at the top of the heap of risk management – the Basle capital rules. If these were done correctly, all else would follow. If there is any regulation at all, it is these rules that should be at the epicentre of global risk constraint and control. But they have failed miserably.

  5. Nemo writes:

    Another excellent piece, Steve.

    What about applying the same reasoning in the context of government response? In the 1930s, the government engaged in massive “economic stimulus”… By hiring people to build roads, bridges, hydroelectric plants, etc.

    How quaint and old-fashioned that looks today! The modern approach is so much simpler: Just send everybody a $600 check so they can buy more crap. Oh, and create hundreds of billions in explicit and implicit subsidies to prop up the prices of purely financial. like houses and MBSes.

    Can’t wait to see how that’ll work out.

  6. VoiceFromTheWilderness writes:

    Excellent rebuttal to what indeed seemed terribly facile to this reader.

    I’d like to add to your point however. We are experiencing more than a crisis of faith in investment vehicles, we are experiencing a breakdown in trustworthy, and well directed leadership, in every area: Political, Financial, Social.

    When the US decidedin late 2001 that it made sense to spend it’s remaining wealth fighting and killing arabs it gave up any hope of actually solving the real cause of our problems: massively increasing dependence on foreign sources of energy and wealth. Our leaders have decided not to invest in wealth producing solutions to our problems but rather in wealth destroying wars that hide the problems.

    If the US were serious about ‘terrorism’, and serious about it’s finances, it wouldn’t be borrowing money from foreigners and the wealthy to pay for wars of aggression. It would be investing in technological solutions to our energy problems using american money. The result would, in keeping with you point, be a reasonable expectation of future economic (and actually political) gains that benefit the american society. Americans are not serious about ‘terrorism’, nor are they serious about the financial strength of their country. They have allowed a select group of people to profit at the expense of the entire society and will reap the rewards of the ‘know-nothing’ ignorance that allowed this decision for decades, if not centuries.

  7. Murph writes:

    This is a thought-provoking article, thank you !

    Basic economic theory posits that lower marginal rates (A) will spur investment, (B) generating economic growth.

    Part (A) worked marvelously over the past several years as lower rates on capital resulted in an investment boom.

    Part (B) failed miserably, as the investment boom was into speculative real estate and related instruments, with no productive value for continued growth. We may as well have been buying tulip-bulbs.

    The lost opportunity form all that misallocated capital is immense. If even half of the ‘investment’ in real-estate items had been into economically productive assets (increased capacity for providing real products and services), we might today be in a sustainable period of expansion.

    I would be interested to know what thoughts anyone has regarding policy prescriptions to limit the chance of another such failure ?

  8. Benign Brodwicz writes:

    Steve –

    Good as far as it goes, but ignores the crux of the current adjustment, I think.

    Most investment is a *derived* demand, from expected demand for a final good.

    Consumption reached 70% of GDP up from its long-term mean of about 66% not on its “merits,” but largely on home equity withdrawals that are basically done.

    So say C drops by 4%-5% of (weakly growing) GDP over the near term. There’s some derived I that will also fall (not just housing, either). So the drop in AgD could be substantial. The tendency toward precautionary saving is only part of the problem.

    Can final demand be engineered to offset these issues? This is the policy challenge for our friends (?) in Congress. Clearly Barack’s income tax proposals are a step in the right direction. (McCain simply lies when he says Obama will raise most folks’ taxes).

    My preference would be for a new New Deal of genuine capital formation infrastructure projects, R&D and education. Some of the bloated military budget (7X-8X that of Russia or China) could be diverted into actually security-enhancing projects in energy research, computers, networking and informatics, etc., as noted above.

    This amounts to government basically taking from the rich, the savers, and directing the proceeds into G configured as much as possible as I, to generate ripple effects. Do I trust government to do this? Not a lot, but the alternative is a depression and the devolution of the USA into a banana republic, IMHO.

    The political resistance to this kind of redesign of the composition of the budget will be intense. Some of my best friends who entered investment banking have turned into hide-bound conservatives ready to let “the market” solve this and all problems. But I think we can kiss a large chunk of consumption demand goodbye for quite a while, and that is the crux of the current dilemma (link to Paul Kasriel’s great piece on this, “It’s So Over for Household Spending”).

  9. Alessandro — I don’t think that the problem is the attitude of savers as much as the reality they face. I think risk-averse potential savers view face sharply negative real returns after you take a haircut for forward-looking uncertainty in the financial system. But as you say, the alphabet soup doesn’t address this issue at all unless it reduces the risk in real terms faced by potential investors. To the degree that people view all the loans and bailouts etc as impacting currency stability, opening the door for future “malinvestment”, or simply being unjust and potentially repeatable, they don’t reduce the uncertainty faced by those who could put their resources into distant future claims.

    Alex R — Excellent points. The boundary between consumption and investment is always blurry, and a lot of our current mess is that we were very good at rebranding as investment what should really have been tallied as consumption. Housing is a case-in-point: Building very large houses doesn’t enable future production anymore than building comfortable but modes homes does, the cost difference should be treated as consumption, but it’s treated as investment. Overpaying for an appreciated home is neither consumption nor investment, it’s just a transfer, but if the transferee views the wealth as a windfall and consume rather than invests the “free money”, at the ends of the chain we end up with “investors” holding mortgages that in aggregate fund consumption. The difference between granite countertops and a plasma TV is very subtle to me, but all that “home improvement” was considered investment.

    JKH — “The US doesn’t save enough to finance the investment it has.” This is right, according to normative data, but see the comment above. We don’t need “more” investment, we need better investment. Much of what is accounted for as investment is in aggregate better viewed as consumption. So far, the US has had no trouble funding its spending, whichever cubbyhole we stick that spending into. Nevertheless, I don’t think present levels of “investment” are a useful baseline. Spending could fall dramatically, including the consumer credit and housing expensitures that get turned into securities, without harming the level of real investment in the US. It’s quality of investment, not quantity, that is really the issue. Domestic savings is low in part because domestic investors know the product is crap (while foreign central banks don’t care, they are motivated by concerns other than return or US economic development).

    We disagree as a matter of definition about where the financial system stops and starts. I think a good financial system is tightly coupled to “outside” economic decisions, and to the degree that an inside/outside boundary arises, that is a symptom of failure. The purpose of the financial system is to guide and incentivize real investment. I think your perspective is more mainstream than mine, and guides a lot of policy, wherein a “Wall Street / Main Street” distinction is commonly made, and the only linkage is “contagion” from crises on Wall Street. Therefore “saving” Wall Street can only help Main Street. But I think that’s wrong. Wall Street and Main Street are the same street except over the very short term, and Wall Street can damage Main Street even while the financial sector is booming. As a matter of fact, it just did.

    Your point about the paradox of thrift and first derivatives is well taken — no financial system will be perfect, and a sudden shift in propensity to save may temporarily be hard to absorb productively. But in this situation, I’m not sure that consumption is a better choice than precautionary storage. Bidding commodity prices up conserves real capital until it can be productively used, while spending risks squandering it. But as you say, this suggests short-term fall in aggregate income. I think we’re getting to the Krugman/Rogoff debate now. There might not be a normative answer, it might be a question of values, tolerating recession in this circumstance enhances future growth but exacts a present cost, and those alternatives have to be weighed. But it obviously depends on magnitudes, a fast recession that to prevent large future costs would surely be worthwhile, and so would spending to avoid a terrible recession at a very low future cost. There are legitimate tradeoffs here, and people can disagree. What troubles me about the “paradox of thrift” reasoning is that people often forget the tradeoff, and assume evading current underutilization is always good.

    Still, the key point is that an effective financial system minimizes the trade-off. Perhaps I’m utopian, but I think it’s quite doable, to come up with a financial system that creates worthwhile projects to match an increased preference for future rather than present consumption. (And I mean a financial system broadly, which might include state components, institutions that enable entrepreneurship, etc.)

  10. Nemo — Good points. I think what we’re learning is that neither our private sector nor public sector institutions allocate economic resources perfectly, and that we need to improve both. But different institutions have different blind spots, and I think diversity is likely to be an important principal for a good financial system, including a mix of private and state investment that swings the pendulum more towards the state than we’ve seen in recent decades. Let’s hope that the bridges and dams and stuff that we build are worthwhile, not “bridges to nowhere”. Quality matters, in government and on Wall Street.

    VoiceFromTheWilderness — I agree with you much more than I’d like to.

    Murph — “I would be interested to know what thoughts anyone has regarding policy prescriptions to limit the chance of another such failure ?”

    The 70-zillion dollar question. I write whiny articles like this ‘cuz they’re easier to write and defend than proposed fixes. But fixes are what really what matters right now. They’re harder, but we should get cracking.

    Benign — Good point on derived demand. Here are a couple of thoughts. 1) “Thrift” isn’t constant, that is a period of greater thrift might but needn’t imply a reduction in future consumption demand. We might (probably should) pursue an infrastructure and tradables investment boom expecting that very high rates of investment are temporary. 2) The difference between C and I really matters a lot less than their sum. If we decide high rates of investment are permanent, we will need to invest today to buld the factories that will build the robots that will build the factories several investment cycles out. 3) A good financial system counters the fall in derived demand by bringing more of the future to the present in terms of what we spend on. If we can invest in 10 year projects rather than at reasonable risk and with assurance that our claims will hold value, we have a lot more to reasonable spend on now.

  11. Benign Brodwicz writes:

    Steve –

    My unhappy inference is that non-market “mechanisms” (i.e., the dreaded “central planning,” at least in the sense of activist tax policy) will be required to counter falling C and I intermediate term.

    When I say “new New Deal” I mean don’t let the government run the project, but use competitive bid contracts, and hope the process doesn’t get too corrupt….

    A cleaner first step might be a revenue-neutral progressive change in personal income tax rates to fund a targeted investment tax credit. This would be especially attractive to folks in partnerships and S-corps and would force some of their precautionary saving directly into investment. One can only hope the money is spent domestically, I suppose. The beauty of bridges and tunnels is that the people working on them have to be here.

  12. BSG writes:

    It never ceases to amaze me how different people react differently to the same thing.

    When I read McCulley’s piece it came across to me as nothing more than the typical PIMCO talking their book. In fact I thought it was more crass than usual in the blatant calling for more transfer of taxpayer funds to support their bets – especially lauding Congress for supporting Paulson’s plan (I once thought we had an official political opposition, I was probably naive.)

    Steve, kudos for another insightful post. It helped me clarify my own thoughts. I also have felt the vibe you described.

    There is something I hope you can clarify in a future post (it maybe too O/T for this thread:) I’m sure you’re aware of the raging debate about the specter of deflation vs. hyper-inflation. I can see how deleveraging can cause deflation as well as how a determined and reckless Fed can, with a few keystrokes, can overwhelm it. While no one can predict, I was hoping you can handicap the prospects.

    Thanks for sharing!

  13. Noni Mausa writes:

    A fine post.

    May I mention another factor — the rather high cost of not saving?

    There’s an advantage to being able to buy large items when a good price crops up, pay bills reliably, and cover unexpected or refundable costs (a health emergency, a car rental, a business trip your company will reimburse you for). With American savings at or below zero, I am sure they are incurring large interest costs, or other non-monetary costs, in addressing the lumpy contingencies of life. The costs of coping with these contingencies is a waste of resources which could be better deployed elsewhere.

    Those costs, I suppose, mostly show up as either profits to banks and credit card companies, or increased pressures on families and friends, leading to real social and health costs of the contingencies which could only be tracked by the most fine-grained analysis.


  14. RueTheDay writes:

    BSG posted: “When I read McCulley’s piece it came across to me as nothing more than the typical PIMCO talking their book.”

    You know, I was starting to get nauseated reading all of the pro-McCulley comments. When I looked at his entry, my first impression was that it was typical of the sort of pseudo economic analysis one expects from an MBA employed by Wall Street. He completely flubbed his explanation of the “Paradox of Thrift” (a term that, near as I can tell, Keynes never actually used).

    Steve posted, in the original entry: “There’s a hidden assumption in the “paradox of thrift” that really ought to get teased out more often. It is true that one person’s spending is another person’s income. But it does not follow that an increase in saving translates to a decrease in aggregate income.”

    This is typical of the sort of confusion that arises from having been exposed to mainstream “Keynesianism” (which has absolutely zero to do with anything Keynes ever wrote or believed).

    There’s actually no hidden assumption whatsoever.

    From p. 63 of my copy of the General Theory:

    Income = value of output = consumption + investment

    Saving = income – consumption

    :. saving = investment

    The real confusion stems from whether income is the dependent or the independent variable.

    From the GT:

    “Let me give examples of what I mean. My contention that for the system as a whole the amount of income which is saved, in the sense that it is not spent on current consumption, is and must necessarily be exactly equal to the amount of net new investment has been considered a paradox and has been the occasion of widespread controversy. The explanation of this is undoubtedly to be found in the fact that this relationship of equality between saving and investment, which necessarily holds good for the system as a whole, does not hold good at all for a particular individual. There is no reason whatever why the new investment for which I am responsible should bear any relation whatever to the amount of my own savings. Qute legitimately we regard an individual’s income as independent of what he himself consumes and invests. But this, I have to point out, should not have led us to overlook the fact that the demand arising out of the consumption and investment of one individual is the source of the incomes of other individuals, so that incomes in general are not independent, quite the contrary, of the disposition of individuals to spend and invest; and since in turn the readiness of individuals to spend and invest depends on their incomes, a relationship is set up between aggregate savings and aggregate investment which can be very easily shown, beyond any possibility of reasonable dispute, to be one of exact and necessary equality. Rightly regarded this is a banale conclusion. But it sets in motion a train of thought from which more substantial matters follow. It is shown that, generally speaking, the actual level of output and employment depends, not on the capacity to produce or on the pre-existing level of incomes, but on the current decisions to produce which depend in turn on current decisions to invest and on present expectations of current and prospective consumption.”

    IOW, for the individual, income is the independent variable; for the system as a whole, income is the dependent variable. There is no “paradox”. As Hyman Minsky once noted, people complain about the lack of microfoundations in macroeconomics when in reality it’s microeconomics that’s missing the foundation – we all learn to construct demand curves by drawing indifference curves tangent to budget constraint lines, but WHERE DOES THE BUDGET CONSTRAINT LINE COME FROM? It’s not an assumption that one simply posits as being “outside the system”, it’s generated BY the system in the form of income.

    Now, back to your point about “an increase in saving translates to a decrease in aggregate income” – that’s certainly not a foregone conclusion, as you point out. Keynes never claimed it was. The effect of an increase in savings will depend on the rate of interest relative to the marginal efficiency of capital.

  15. JKH writes:

    For my part, I assume most people talk their book, whether it’s a financial commentator with a personal or professional portfolio, or a writer doing a road show, or a blogger channelling in real time a yet unwritten comprehensive world view. I also assume they are not self-defeating and that their book aligns with their talk. I look for quality in both the talk and the book, and try not to let my assessment of quality be too biased by its philosophical alignment, although this is hard. McCulley seems to me to be a cogent analyst of the Fed, interest rates, and related economics. He understands particularly well the actual plumbing and detailed effect of Fed balance sheet operations. The quality of the analytical insight is rare, notwithstanding whatever one’s attitude might be toward “Keynesianism”. Conversely, stealth conveyance of one’s book alone can curdle in an unseemly way with personal agenda.

    I’m not too concerned about the precise explanation of the paradox of thrift. The conversion of foregone consumption to longer duration inventory investment is what ensures the persistence of the identity of saving and investment on a continuous basis. I agree that McCulley’s summarization of the paradox of thrift is somewhat sloppy. But whether he flubbed it or just skipped a bunch of steps, it is very far from the main point of his essay. Extending the notion to financial leverage is the far more interesting and relevant idea here.

    P.S. This is one of the very best economic blogs. Not only is the writing unusually penetrating and thought-provoking, but as importantly, it is seamless in its high quality as between the original post and feedback on comments. Thanks for the continuous investment. :)

    And I look forward to your upcoming scheduled guest blogger appearance on naked capitalism. Are you prepared for a likely spike in traffic here as a result?

  16. RueTheDay writes:

    One further point. Here are the parts of McCulley’s piece with which I strongly disagree:

    “But monetary easing is of limited value in breaking the paradox of deleveraging if levered lenders are collectively destroying their collective net worth. What is needed instead is for somebody to lever up and take on the assets being shed by those deleveraging. It really is that simple.”


    “But levering up Uncle Sam’s balance sheet, to buy assets to break asset deflation resulting from the paradox of deleveraging still seems to be a foreign, if not a sinful proposition. This need not be, and should not be.”

    Theoretically, the price of a capital asset should equal the net present value of the discounted set of income flows the capital asset is anticipated to yield. I say “anticipated”, because no one can predict the future.

    If the market is overreacting and pricing those assets below the level indicated by realistic estimates of their future income yields, then McCulley has a point. However, if the only “justification”, and I use the term loosely, of those prices was a leverage-induced speculative mania predicated on wildly unrealistic expectations of both future income streams, and, more importantly, expectation of unlimited future asset price gains, then McCulley’s solution is a non-starter. I lean strongly towards the latter view.

    Would McCulley have advised the federal government to borrow enormous sums to buy up dotcoms in late 2000 in order to prevent the unfolding deflation in asset values in that sector? I would certainly hope not, so why is he doing it in this case?

  17. BSG writes:

    JKH – your point is well taken. I should clarify: what I find particularly repulsive is the lobbying for taxpayer/Fed money in the guise of analysis, sometimes accompanied by outright fear mongering. Some of the PIMCO pieces have provided reasonable analysis even while talking their book.

    It occurred to me before reading Steve’s post that McCulley’s piece started with a lobbying purpose with the analysis (possibly deliberately) tailored accordingly.

    It’s not a crime. It’s just that it grows tiresome and in my view should be discouraged.

  18. BSG writes:

    RTD – you expressed my thoughts better than I could. Your 2000 dotcom analogy is a good one and I would venture a guess that if that was his primary book, he would have. Possibly rationalizing that retirement savings and/or tech leadership were at stake. Of course, we can’t know that – it’s just to illustrate the point.

    What makes it much worse is the level of devastation the fiscal and (at least prospectively) monetary crisis that seems increasingly inevitable as the US government proceeds on the current course.

    I sense I may be beating a dead horse, so I’ll stop.

  19. JKH writes:

    If you have a broken and badly functioning financial system, you want the pilots and the controllers to attempt the controlled crash landing of a rather large vehicle. You don’t want them to immediately eject and abandon stations for the redesign meeting, while the existing system proceeds to cartwheel into the corn field of the real economy.

  20. Nels Nelson writes:

    At some time in the recent past, Bill Gross sent one of his minions out into the U.S to determine if a real estate bubble was forming and if there was one forming to what extent. The report came back that there was indeed a bubble. The bubble was inconcealable but the consequences of the bubble were not discussed.

    Brad Delong, Larry Summers, Bob Rubin and their like from the growth at any cost club defended Greenspan’s serial bubble blowing because it was their belief that the eventual collapse could be ameliorated and the growth resulting from the bubble would be greater than the contraction after the collapse. In essence a net gain and ultimately we’re all better off.

    Few of the financial system’s guardians gave us any warning of the implications of the bubble and those that did were ridiculed and intimidated. Like Admiral Farragut it was “damn the torpedoes, full speed ahead.” Now the bubble defenders all profess to be surprised by the magnitude of the bubble and its consequences. These consequences are too dire to comprehend and therefore justify taxpayer funded bailouts.

    It was stated somewhere in a previous post that what we’re suffering from is a crisis of leadership. At opposite ends of a spectrum we have two highly persuasive positions: central planning and free markets. What we need is leadership to lead us along the critical path between these opposites. What we’re dealing with are massive egos who believed they could fine tune things. They need to be publicly called out and held accountable.

    In analyzing the unfolding events, a question in my mind that needs answering is: are we a maturing economy that is losing its ability to produce and create value? This would seem to be the case given the rise of speculation as a replacement for real industries and jobs that have been destroyed.

  21. nextinline writes:

    Not sure I get the logic of the piece on the “Paradox of thrift”. If people buy “gold, or oil, or art, or whatever”, don’t the people who sold them that “gold, or oil, or art” earn income which they can then spend/invest? Exactly where does the pipe spring a leak? Again, I remember reading somewhere that, historically, consumption is fairly stable during a recession, but investment falls off. Which seems to imply that businesses are holding back investment even as households are consuming away. Still don’t know where the leak is..

  22. Great post, I very much appreciate the logic here. This is a timely consideration as well.

  23. JKH writes:


    “Not sure I get the logic of the piece on the “Paradox of thrift”. If people buy “gold, or oil, or art, or whatever”, don’t the people who sold them that “gold, or oil, or art” earn income which they can then spend/invest?”

    I have an indirect comment without relating directly it to the Paradox of Thrift.

    It depends on whether the gold, oil, or art is newly produced and considered part of current GDP (the way it is actually measured, or conceptually revised if one wishes), or something produced earlier. (Oil tends to be “new” in this sense).

    If newly produced and part of current GDP, we can assume both the producer and the buyer have earned income. The buyer has purchased the producer’s output with his income. Somebody else will purchase the buyer’s output with their income. The producer will purchase something else or save with their income. The current output and income markets will clear.

    On the other hand, if the gold, oil, or art is “old”, it has essentially become an aged asset rather than a new product output. In this case, the transaction is essentially independent of current GDP and the income and purchasing power associated with current GDP. In accounting terms, it is a balance sheet transaction but not an income transaction (at least on a net basis). Two assets are being exchanged – money and the other. And if the money is actually or selectively associated with the income result of current GDP, the transaction effectively only transfers the ownership of that income result and its purchasing power from buyer to seller. But it doesn’t match up current income with current output, which must be done before everything in current GDP gets cleared.

    Again, this interpretation has little to do with the technical definition of GDP and whether one agrees or disagrees with it. The point is that one must draw a demarcation line somewhere between new and old when discussing issues of economic output and income.

    The logical implications of distinguishing between a concept of new and old are the source of a lot of confusion in economic analysis, in my opinion.

    The “old” piece is sometimes called the asset economy. Stephen Roach of Morgan Stanley (formerly chief economist; now head of Asia) has written great things about the deleterious impact of inflation in the US asset economy. A good part of the housing crisis has impacted GDP negatively (output in the form of new housing), but an even greater portion has brought down values and wealth in the asset economy (existing housing). And a corresponding amount of the mortgage and financial problem is related to prior asset inflation rather than GDP inflation per se.

    Stephen Roach has received enormous criticism over the years for being a bad forecaster in terms of timing. But he has been brilliant at seeing the nature of the train wreck at the end of the tunnel and writing about it.

  24. nextinline — It’s often useful to take money out of the picture to help think about things. Suppose you have gold and I have oil, and we trade some. We may both be better off in some sense (I wanted more gold in my portfolio and you wanted more oil, for whatever reasons), but if we look through individuals to the aggregate economy, nothing has changed. There is still just as much gold as there ever was, and just as much oil. In the aggregate, there has been no “income”.

    Add money to the mix. I have dollars in a bank account, you have gold, we trade. Again, nothing has happened in the aggregate. Tautologically, you have earned a an “income” in dollars (and I have spent some money), but the transaction resulted in no new goods and services being made available to the aggregate economy. Our transaction hasn’t added to GDP (unless there was some intermediary, who was paid for the service of getting us together, whose commission would be considered income).

    When speaking of aggregate income, we try to exclude transfers, and put a value on the new goods and services we’ve produced. So paying you to give me gold “adds a lot less value” than paying you to turn sand into microchips.

    Of course all of this is muddy and sketchy. If we are really both better off by virtue of a portfolio shift, and would even have paid a broker some of that surplus, oughtn’t what we would have been willing to pay to find one another be counted as “income”, even if we did it ourselves? And, of course, shouldn’t we take into account the goods and services we’ve used or destroyed, not just those we’ve produced, and think about a “net domestic product” or somesuch? There are a lot of tricky issues in thinking about aggregate income. The intuition of ordinary, mainstream GDP is that the health and vibrance of an economy is related to activity and new production more than anything else. That’s arguable on a lot of levels. But it’s not incoherent. And, from this perspective, not all transactions are created equal. We can quibble about the details, but some transactions approximate mere transfers with little or no value added, while some transactions are endow new production almost completely. Holding the dollar value of transactions constant, mere transfers represent a kind of leakage from an aggregate income perspective. If everyone takes their money, optimizes a portfolio of preexisting commodities, and holds, very little production or aggregate income ensues. We are just dividing up a preexisting pie, and working to store or preserve it.

    (If we wish to hold corn and are paying people to produce it, that’s a different story. If we wish to hold oil and pay people to deliver it from under their land, it’s a mix, partly a mere transfer of stored product, but part real production as oil in tanks is a much more useful and convenient product than oil beneath the seabed. Again, there’s a continuum, not a bright line, between mere transfer and new production.)

  25. JKH — Great minds think alike (and sometimes even fools like me get it right!). Looks like we were simultaneously writing almost exactly the same thing.

    Thanks also for the kind words. I love your airplane analogy above. An important point, and a nice analogy to think things through with. We want as gentle a crash landing as possible on the one hand, but we don’t want the bastards running the airline to stick us back on the same broken plane for our next trip, pointing out mildly that sure, there was a bit of turbulence but we did manage to get through it, didn’t we?

  26. RTD — Following your example, I finally do have hardcopy of the General Theory, of which I’ve only begun a linear read. I did dip into it on thrift, and as you say, I found no mention of the phrase “paradox of thrift”, but a description of an already long-standing debate on the topic of whether and when profligacy might be good for an economy. Keynes’ position struck me as nuanced, not inconsistent with the piece I was preparing to write, but also not inconsistent with the position often attributed to him that under some circumstances encouraging consumption might have useful effect. It’s just that the “under some circumstances” is usually left out. The common distillation that whenever GDP is sputtering, we should stimulate consumption, does not strike me as a fair read of Keynes position. Were such a question put to him, he’d probably respond “it depends” (which I think is the right answer). As you (and the paragraph you quote) suggest, it depends on the perceived (and eventually real) tradeoff between production and storage (which is really just the residual part of investment not put into active use towards future production). If people perceive the real interest rate to be higher than the marginal productivity of capital or the expected appreciation of a marginal unit stored, they’ll save in banks, who will have a problem. Either the interest rate they pay will have to fall in nominal terms, or they’ll invest in underpriced government securities leading eventually to inflation as interest is paid faster than available goods and service increase.

  27. Nels — I’m broadly optimistic, in that I don’t think economies are like individual industries, destined to mature and then slow. Humans are too good at manuafcturing and appreciating novelty for that.

    But I think you’re right on in discussing the tension between central planning and allegedly decentralized markets, and we also need to think about the structure and quality of each, not just set them up as iconic extremes. The market part of our system has a good deal of power politics and central planning, and the government part, while very deeply flawed, has different blindspots than our market institutions, and shouldn’t be ideologically illegitimized. Our institutions on both sides of the public/private divide are very deeply flawed, and the best approach to fixing things is not to set up some Olympian Objectivist struggle between capital and the state, but improve the decision quality of both sets of institutions and define some equilibrium where each “checks and balances” the other. Much easier said than done, sure, but so is getting out of bed, and we do sometimes manage it.

  28. Noni — Nice point, hearkens back to the convenience yield conversation. There’s a big convenience yield associated with holding cash, and during a credit crunch, that yield is available exclusively to savers!

    BSG — I’d like to write more about the great inflation/deflation debate. I think it’s a hard question, though I’m lean towards inflation, for reasons summarized here.

  29. BSG writes:

    Steve – re. the airplane analogy, I recall a post of yours from a while back in which you wrote that central banks are dangerous. I think it was in the context either of the BS bailout or the Fed loosening the collateral standards.

    The impression I got was that you did not approve.

    I am wondering whether you have reconsidered and now consider the powers that be as reasonably piloting a plane in trouble, notwithstanding the obvious risks of crashing into a heavily populated area in the process of trying to land (to expand the analogy, though I do think it has its limitations.)

    Although we’ll probably never know, it seems to me that aggressive management of the crisis without bailing out so many bad or reckless actors stood an equal or better chance of working, and still does.

    If the steps taken so far continue at every turn, I don’t see how we can avoid a “full faith and credit” crisis, which would obviously be worse.

    To your point about being put back on the same plane, with the current approach I don’t see how that can be avoided if we get lucky and the airline crows that it was just some turbulence and they had it under control all the time.

    This is probably a failure of imagination on my part, but it seems to me that if strong medicine is ever called for, the sooner the better. I don’t see how we can get lucky forever.

    I would appreciate whatever insight you can offer.

  30. JKH writes:

    Speaking of new and old, I have a brand new copy of Keyne’s “General Theory” that a friend purchased for me 4 years ago. I have yet to read it. But I’ve read enough excerpts over the years to come to the opinion that Keynes is the James Joyce of economists. It’s not the policy derivation of his writing, but the thought and the analysis that is simply overpowering in its pure density and excellence. I also have a brand new copy of Joyce’s “Ulysses” that I purchased more than 30 year ago. I’ve never been able to get past the first page without being blown away. (A university professor I once had, widely considered my own country’s most accomplished fiction writer, said he had never known anybody that had read Ulysses in its entirety). Both of these books await the right moment. (Paul McCulley is no Keynes or Joyce, but there aren’t that many that can write as naturally and accurately about the operational interface between the Fed and the free market the way he does.)

  31. cato writes:

    good analysis…the problem is the real savers need a real unit of account in order to invest in something real and worthwhile…

    hence gold, which is imperfect, but don’t let the good be the enemy of the perfect…

  32. Blissex writes:

    «Part (B) failed miserably, as the investment boom was into speculative real estate and related instruments, with no productive value for continued growth. We may as well have been buying tulip-bulbs.»

    If only it were so! A lot of the money borrowed at way below zero real rates has been used to fund a colossal amount of capital investment in China and India, donating them in effect a whole new industrial infrastructure to compete with the older, less productive one in the USA.

    The advantage of the USA economy was a much higher capital endowment than China or India, supporting higher productivity and wages. But the USA government have been frantically trying to ensure that this advantage disappeared by funding a lot of investment abroad at very low cost.

    The discussion above, even the excellent quotes from Keynes, are missing this point: that savings and investment do not necessarily happen in the same country, they are not just made by different people.

    So extremely loose cost of financial capital in the USA does not necessarily lead to a boom in productive investments and jobs in the USA itself, but in China and India, which are relatively job and productive asset poor.

  33. Blissex writes:

    «Brad Delong, Larry Summers, Bob Rubin and their like from the growth at any cost club defended Greenspan’s serial bubble blowing because it was their belief that the eventual collapse could be ameliorated and the growth resulting from the bubble would be greater than the contraction after the collapse. In essence a net gain and ultimately we’re all better off.»

    The Greenspan/Bernanke/Delong/Summers/Rubin/… side is the side of people who know very well that the USA are running two very expensive, unpopular wars (and a half), and the good times must roll, regardless, both to keep the home front happy and not thinking about those wars, and to finance those wars with cheap debt.

    Just like most people forget that financial capital is highly mobile, and loose money in the USA can well result in jobs, investment and inflation booms elsewhere, most people forget that the USA are at war, and the government is running a war economy, and an internal propaganda war. The various wars that the USA government are waging are the first in the history of the world entirely financed with tax cuts after all.

    As long as people can see the prices of the assets they own go up, they will think “F*ck you, I am fully vested” and not worry a bit about the war — why worry, be happy? This is elementary management of home front morale.

    In effect a very large mass of the USA electorate (most of which are asset owners) have been turned into war speculators.

  34. zanon writes:

    As always, a great discussion.

    One point I’d like to bring out is how, in the General Theory, as well as more “micro-based” models, there is no identity for squandered investment, or waste, which I think goes to the heart of your point, Steve.

    Income = value of output = consumption + investment

    Saving = income – consumption

    :. saving = investment

    DeLong, Krugman, and other professional economists always see stimulus as being a good thing because all activity is necc. good. I agree with you Keynes would say that “taking tax money to pay people to dig holes and fill them” is a good strategy, but only in extremis. Extremis today seems to be whenever the economy slows even slightly.

    General Theory, IS IM, and AG AC models don’t have any element for waste, and cannot model bubbles. There is no

    Value of output = consumption + investment + bubble

    To this end, any period where output growth was goosed by unsustainable bubble activity cannot be followed by a period where output growth occurs at a lower rate *without* the bubble goosing, and even worse, off a lower base now that there is less capital available because some of it was squandered (a lower income constraint, in micro terms).

    A saver who wishes for a quiet life is at a very difficult place in the US. If he keeps his money in $ in the bank, the government might take it via monetary dilution (inflation) or some form of higher taxes by running up every larger deficits. He might be forced to spend it, but this forced spending would likely be moving consumption forward (which is very hard to do actually) or, more likely, try a new form of saving just to preserve wealth (ie buy oil, gold, real estate, etc.)

    While the distinction between investment and consumption is clear in equations, you are right Steve, it is much fuzzier in real life. The $750K McMansion that you took a neg-am loan to buy — was that investment? or consumption? or both? If a portion of the “investment” piece goes to zero, do we reclassify it as “consumption”?

    The fact that the US economy seems incapable of deleveraging safely is, frankly, a total failure. It means that money supply can always increase, by extending credit, but the only way to recover from bubbles is by transferring wealth from those who are net long to those who are net short. This is not an environment that encourages sensible real investment decisions.


  35. BSG — I still don’t approve. My bottom line is that those piloting the crash landing need to be very publically and non-repudiably admitting that they are in fact piloting through a crash landing. What bothers me about the present strategy, which I do on balance oppose, is not that it is trying to limit “collateral damage” and harm to the real economy, but that it is trying to retain the financial system status quo as much as possible, by acting as covertly as possible and minimizing and perception of discontinuity. As you suggest, banks could be nationalized rather than bailed out, cooperation by organizations could be compelled with sticks rather than with carrots (financial organization principals could be put in civil or criminal jeopardy for failing to minimize disruption during transitions the terms of which they may not approve). I am not at all happy that on balance, the financial community has been net rewarded for its idiocy this decade, rather than net punished, and that most individuals who participated in the boom, should circumstances repeat, would be better off doing exactly what they did for a while (perhaps retiring a bit sooner) than doing anything different. We can’t undo people’s past salaries and bonuses, but we can make them not worth the hazard in retrospect on a risk adjusted basis. That requires creating hazard, which we have shirked from doing.

    All of this does nothing to invalidate JKH’s point that we do have to get this plane down to the ground with most of the passengers unharmed. My main bottom line is that it had better be clear when we stumble off that the contraption was badly put together by people who profited from cutting corners, and we’d better do something about that before we fly again (and we must always fly again).

  36. BSG writes:

    Steve – very well put! I also agree with JKH on that. I just think that the sort of measures you alluded to could and, more importantly, still can, serve both to facilitate the “landing” and to prevent future disasters.

    Thanks for a great analysis.

  37. Blissex — I’m sympathetic to some of your points, and think it especially important to consider the international dimensions of things. In the context of this discussion, I think the international dimension is complicated. We had kind of an antiparadox of thrift, where supposed investment was really devoted to present consumption, including a very great deal of foreign investment. It’s not obvious why squandered foreign investment would result in an apparent capital deepening within the countries whose capital was squandered, and a capital “shallowing” in the capital recipient who did the squandering, but I think you’re right that something like that happened. But we’d have to talk about more than just quantities of “C” and “I” to get at why, we’d have to look at the qualitative dimension of those quantities, what is being invested where, and why, to see why it may have been in the interest (from a national power and industrial development perspective) for some nations to let their capital be squandered, and why an overabundance of capital in the United States failed to translate into a sustainable capital deepening, but were devoted to consumption or overinvestment is sectors that predictable would be required to shrink when the flow of foreign capital ceased. I do think that’s what went on, but we can’t understand that in terms of quantities of consumption and investment, in an accounting sense the US had an investment boom, it’s just that much of that “investment” really belongs in Zanon’s “bubble” term.

    Zanon — I love your discussion, and agree with nearly all of what you have to say. It is a common error among economists to assume that whatever it is you aren’t studying will take care of itself. If you are not investigating market efficiency, you assume markets are efficient, which provides for well-defined behavior so you can make sense of whatever you are investigating. If you are not investigating bubbles or misallocations of capital, you presume that a scalar variable I is sufficient to describe investment, because despite the variety of possible ways real diverse stuff might be used as capital, there is only one “optimal” mix for any quantity, and efficient markets would certainly find it, so the quantity is the only free parameter. Of course that’s total bullshit, but it does simplify the modeling. As you say, at the very least, we need to include a quantity for misallocation or waste, but once you allow for that, how do you quantify it? Sensible people knew it was “a lot” in the recent past, but if you’re an economist, how do you justify an estimate of “40% wasted” before any losses are tallied? For modeling purposes, zero was always the most defensible number. Given the large quantity of “investment”, with zero waste and presuming modest to normal returns, things looked just peachy. Until they didn’t. Sometimes the most defensible model is dead wrong. Quite often in fact.

    I like your point about deleveraging. Ironically, if “investment” had actually been investment and it had simply failed, the “deleveraging” would be easier, because we would have lost an expected increase in consumption before we’d hit current consumption. When we have leveraged present consumption, deleveraging hurts, because it means a diminishment of our existing lifestyle. As you say, our difficulties in managing deleveraging says something not so nice about how forward-looking our use of “capital” has actually been.

  38. Benign Brodwicz writes:

    Nice perspective on US total credit market debt / GDP and the long-term cyclical position of the US stock market:

    Ronnie Reagan started us down the primrose debt-bubble path with government borrowing to cover up the short-comings of “supply-side economics” (it was really “mourning in America,” mourning the end of the post-war American boom and debt-binging to get over it)–and the consumers and firms just followed along. Supply-side economics was just a ruse to cut taxes on rich people.

    Although the chart doesn’t show it, the last time total credit market debt / GDP hit these levels was in 1929.

  39. McCulley’s article might be an easy read, but it seems like rubbish to me. I cannot believe, for example, that the Fed would call the vicious circle dynamics he describes as a “negative feedback loop”.

    I can appreciate that some cushioning of the fall in asset prices may be desirable, although its aim should be to reduce the gradient not the size of the fall. But the primary role of the authorities in this should be to organise it. First, do not stand in the way of any shareholder wipeouts. If a failing institution is too big to fail, forcibly nationalise it, break it up, and sell the bits. Second, unless there is evidence of predatory lending, do not spend taxpayers money to help defaulting mortgage borrowers to stay in their houses – give them an adjustment grant to move into good rented accommodation, and give a marginal homeless person a grant to help them take over the house. Third, to the extent that taxpayers’ money is required, raise taxes as close as possible to those responsible for this mess. We hear about windfall taxes on oil companies, so why not a levvy a windfall wealth tax on anyone working in the finance industry whose tax returns showed that they earned more than, say $10mn over the last five years?

    I never really understood the paradox of thrift, because I was not convinced that prices would not adjust to break it, but if anyone knows of a convincing explanation, please tell. However, it seems clear to me that the USA has a capital shortage. Given its ageing population, America was probably not saving enough even given the unrealistic expectations of return on investment that had been made until a year ago. But now that these expectations have been falsified, it seems to me that some serious investment is in order. Output will rise as the capital stock grows, but not consumption, as much output will have to be ploughed back into increasing the capital stock further and much output will have to be exported to pay foreign creditors, so it will feel like a recession anyway.

  40. a writes:

    3:11 p.m. rude late comer:

    Apparently our host thought McCulley’s piece was worthwhile enough to be “thought-provoking”. He developed an impressive post from that starting point. And many here appreciated the quality of the lengthy discussion that resulted.

    Would you appreciate somebody advertising the work on your own blog as “rubbish”? Be more careful with your words and your manners. You need a little more of a following to be quite so pretentious. And in the interim, why would any one bother to explain the paradox of thrift to you?

  41. a

    Perhaps I was a little too dismissive, but I do get irritated when PIMCO’s gimmicky, affable notes get attention for what seems to be talking their own book – in this case a plea for public money to restore confidence in spread product. But please do advertise the work on my blog in any way you please – all publicity is good publicity :)

  42. shrek writes:


    I dont think were going to see the bottom or a return to financial health until serious punishment are reforms are undertaken. The good news is the chances get higher after november.


  43. By the way, the reason why I think that it is vital to have a rigorous understanding of the paradox of thrift (McCulley’s article was really about another fallacy of composition which he compared to the paradox of thrift) is that the paradox of thrift is used to make the argument that a unilateral increase in the US savings rate would cause a global slump (see Martin Wolf on his blog and on the back page of Monday’s FT). I am not convinced, although it seems reasonable that there might be an adjustment period of reduced output as some production of consumption goods and services was shifted into investment goods. Presumably, if there is no paradox of thrift, the present policy of monetary and fiscal stimulus is inappropriate. The discussion above did not seem reach a conclusion about this, although I appreciated the quotes from the original Keynes. I was hoping that someone might know of a better account of it since Keynes (eg Leijonhufvud?).

  44. a writes:

    4:49 p.m.:

    Here’s the Wiki article on it. It actually makes a reference to you in it, and categorizes you according to your own misunderstanding, which in a lovely way is the opposite of what you contend above the article. See the sentence beginning “One who does not know …”

    It basically says everything you deny about it, but since you’re bent on inventing your own definition, do it, and then put it up on Wiki. Enlighten the world with some supra Wiki intelligence and sophistication! And then advertise like crazy!

    Paradox of thrift

    From Wikipedia, the free encyclopedia

    The paradox of thrift (or Paradox of Saving) is a paradox of economics propounded by John Maynard Keynes. The paradox states that if everyone saves more money during times of recession, then aggregate demand will fall and will in turn lower total savings in the population. One can argue that if everyone saves, then there is a decrease in consumption which leads to a fall in aggregate demand and thus leads to a fall in economic growth.

    The simplified form of the argument is that, in equilibrium, total income (and thus demand) must equal total output, and that total investment must equal total saving. Assuming that saving rises faster as a function of income than the relationship between investment and output, an increase in the marginal propensity to save, all other things being equal, will move the equilibrium point at which income equals output and investment equals savings to lower values.

    In this form it is a prisoner’s dilemma as saving is beneficial to each individual but on a whole it can be harmful. This is a “paradox” because it runs contrary to common intuition. One who does not know about the paradox of thrift would fall into the fallacy of composition. This fallacy arises when one infers that something is true of an economy from the fact that it is true of an individual. Although exercising thrift might be good for an individual, by enabling that individual to save for a “rainy day”, it might not be good for the economy as a whole.

    This paradox can be explained by analyzing increased savings in an economy. If a population saves more money (that is the marginal propensity to save increases across all income levels), then total revenues for companies will decline. This decrease in economic growth means fewer raises and perhaps downsizing. Eventually the population’s total savings have remained the same or even declined because of lower incomes and a weaker economy. This paradox is based on the proposition, put forth in Keynesian economics, that many economic downturns are demand based.

  45. a

    I am not sure what you mean by “it….makes reference to you”. I have not written anything about the paradox of thrift, because I am remain open-minded about it, but sceptical since it is, after all, a “paradox”. What I am sure about though is that, often, ideas about saving are flawed because they do not account for its counterpart. It seems to me that to save I must either persuade someone else to borrow, or buy some kind of investment item. I can understand that certain kinds of saving are inefficient (hoarding fuel, probably), and if widely practiced due to a severe loss of confidence can lead to a slump, but that is not the same as saying that saving per se is contractionary.

  46. a writes:

    6:36 a.m.:

    For starters, saving has nothing to do with borrowing per se. This is its own type of fallacy. For example, businesses save and invest without borrowing.

    Second, if you choose not to spend, the item on the shelf inadvertantly becomes a protracted inventory investment for business. Businesses retrench their full production and investment processes as a result. Excess saving leads to contraction.

  47. a writes:

    And you don’t have to persuade anybody else to do anything in order to save. You simply withhold from spending your income. Full stop.

  48. a

    Allow me to try to convince you:

    One of the counterparts to saving I mentioned was to “buy some kind of investment item”. As you say, inventories count as investment. Of course, a firm does not purchase inventory from itself, but it must pay its production costs for that inventory, and if it chooses not to sell the inventory, it must raise the money to pay its costs somewhere else. The firm must either run down its own savings or borrow, in which case someone else in the economy must save.

    If I just stop spending my income, what happens to it? If my income is paid into a bank deposit, then the bank will accumulate some asset – ie lend to someone. If I retain my income in banknotes, I have chosen to hold more of a marketable loan to the central bank / government. Banknotes are just bearer debt securities.

  49. a writes:

    9 a.m.:

    We’ll have to disagree. I’m not sure what you’re saying, but a recap of part of what I’ve said:

    Inventories of consumer goods are an output of GDP. They have already been produced.

    As an output of GDP, they must generate an equivalent amount of income for factors of production.

    This factor income is equivalent in size to a specific amount of income somewhere in the economy that cannot yet have been expended on the purchase of these consumer goods, since they’ve been produced (resulting in income) but not yet sold to the consumer (i.e. not yet resulting in expenditure). Therefore, someone in the economy must be saving an amount of income corresponding to this output. Any other conclusion is a logical contradiction. This amount of income must have been saved in order to fund the same amount of inventories in the economy.

    So the starting point for inventories at any level is an equivalence of savings and corresponding investment, as a subset of all saving and investment in the economy. Expenditure on consumer goods is then an ongoing transition from the classification of inventory investment to the classification of consumption. This is happening all the time. The paradox of thrift suggests that the attempt by consumers to save by withholding the expenditure process will temporarily cause inventory investment (and saving) to become bloated, which causes businesses to retrench, which causes both inventory investment and long term investment to decline, which causes incomes and savings to decline, etc. etc.

    The required financial intermediation for inventory investment and anything else is irrelevant to the economic result. No amount of circuitous permutation of finance can change what happens according to the underlying economics.

  50. a

    I would be sorry to agree to disagree. As someone who took a career in science as far as a PhD, this is something I find frustrating about economics. In a science, the two sides would refine the point of their disagreement in an attempt to identify a key proposition, test it either by experiment or logic, resolve the disagreement and move on.

    If you take as your starting point a situation in which the inventory has been produced, and the initial act is that of the consumer to retrench, then the consequence of them not spending (saving) is that the firm does not receive income it had expected to receive. Initially, this shock forces the firm to either save less than it would have done, or borrow to fund its unsold inventory (negative saving in either case), so there is no net saving. Thereafter, it is not clear what happens to output, unemployment etc, because this depends on, first, the effect of the shock on interest rates, second, the response of the firm to that change in interest rates (ie is it happy to hold higher inventory or does it cut production to work off inventory), third, the effect of any production cuts on product prices, wages etc, then the response of the economy to those changes, and so on. And even if output does fall, this is not necessarily undesirable or wasteful, as, assuming that the markets are clearing, this is consistent with the modified preferences of consumers. Ultimately, it comes down to the debate between real business cycle theorists and new-Keynesians about whether markets are clearing, and as far as I know, that debate continues.

    The practical point of all this is, is the amount of saving that the US (and UK for that matter) has been doing reasonable? My guess is that there has been too much borrowing based on unreasonable expectations about returns. If so, in the absence of clear evidence that markets are not clearing, I would err on the side of allowing saving to rise as the return mistake is recognised, rather than, for example, attempting to forestall adjustment by debt financed government spending.

  51. a writes:

    “so there is no net saving”

    Not so. As I explained, existing inventory is financed by saving somewhere in the economy, by definition. The same holds for new and even unwanted inventory, since factors of production have already been paid for the output, but a corresponding expenditure hasn’t been made at the macro level. So it must be financed by saving from somewhere.

    But more generally, I think you’re not addressing the paradox of thrift – rather you’re effectively debating whether or not the preconditions for it (attempted “excessive” saving) can revert due to changes in other variables like monetary policy or knock on effects like changes in business pricing. I suppose anything you want to assume as possible is possible. But I don’t see how the core of the analysis has meaning without considering marginal effect first, before assuming all sorts of reaction functions to changes that occur beyond the first order change. After all, the first marginal effects are what can lead to things like changes in monetary policy or changes in business pricing.

    Likewise, it’s obvious that US authorities are trying to prevent an overly abrupt slowdown via monetary and fiscal policy. This is consistent with preventing too abrupt and destabilizing an adjustment in the US savings rate – even if only from horribly low to somewhat better than horribly low. Clearly, the US needs a much higher savings rate and better investment strategy on a secular basis. The question is how quickly can the economy adjust toward that at this time, given current financial market conditions, without imploding? But I’m not debating the wisdom of policy prescriptions here. I’m just looking at first order analytic differences, as in the first paragraph above, for example.

    (On policy, for what it’s worth, I’m in favour of a “measured pace” (to use Greenspan’s phrase) in moving the savings rate higher and the current account deficit lower. So I’m somewhat sympathetic to monetary and fiscal cushioning in the rate of such necessary changes. But I generally agree with the direction of your final preferred outcome.)

  52. Sorry for my sloppy use of the word “saving”. I meant to say that the reduction in consumption was matched by an increase in investment, so that aggregate demand was unaffected (at the first round). Like I say, I am prepared to be convinced, but I cannot see any way of saving that reduces aggregate demand at the first round. There has to be an outlet for saved resources somewhere, either in investment or in someone else willing to purchase your present resources (for their own investment or present consumption) in return for a promise of future repayment.

  53. a writes:

    I agree that the reduction in consumption expenditure and attempted increase in saving results in a matching increase in inventory investment. And therefore aggregate demand is still unchanged at that point. But elevated inventory levels are now out of synch with final demand. And since the paradox of thrift assumes a persistent attempt to increase saving, inventory levels will only grow further at prevailing levels of production and consumption. This can only be corrected by a reduction in production, income, and macroeconomic saving levels (relative to intended saving levels.)

  54. a

    Hopefully, we now have a meeting of minds on the effect of the initial retraction of consumption (although it is worth noting that my argument is based on markets clearing and resources not being wasted – eg involuntary unemployment – which Keynes effectively disputed). The resources saved by cutting consumption initially reside in inventory investment.

    So now it is necessary to consider how the firm responds – which I referred to as the second round, which is less clear. Perhaps you will find my conclusion challenging! There are links between our discussion and that on the following thread, especially the comment from JKH, who I know from other discussions is a smart guy. As JKH says, for the system as a whole, inside (ie financial, IOU etc) saving nets out. Any saving must be outside saving, meaning real investment. In our case, in the first round the saving has gone into inventory, held by firms. Ultimately, however, trading based on prices including interest rates, should allocate the desired outside saving (investment) to the best place (eg according to return and risk).

    After round 1, the firm has higher inventory and a smaller bank balance than it expected. Consumers, who started the process, have a larger bank balance. It seems likely that consumers’ saving initiative has led to lower interest rates. So the firm has to decide whether it is worthwhile carrying higher inventory at lower interest rates. As I said before, it is conceivable that they do in which case round 1 is sustainable, but unlikely, especially if their inventory is perishable. The firm is therefore likely to try to reduce its inventory by a combination of lowering its output prices and cutting production. Round 3 is the consumers’ response to changed prices and changed labour demand. And then the firm responds to those, and so on.

    Where the new equilibrium ends up after the shock to consumer preferences is difficult to say. It depends on how prices change and how participants in the economy respond to those prices. But I ask you to make one leap of faith. That is that consumers’ preference for increased saving ends up with an increase in investment, with the increase having a positive return. Essentially, assume that the economy is able to adjust to find some productive use for the increased outside saving.

    So far, this has been a story about flows. However, while depreciation does mean that investment flows do not just accumulate ever more capital stock, an increased flow of investment will produce a higher capital stock (ie the bucket may have a hole, but pouring in water faster will increase the level of water in the bucket). Now, if resources are not being wasted, and the capital stock is higher, the output of the economy will be HIGHER when more is saved. In short, assuming that resources are not being wasted, thrift does indeed lead to higher output – in that sense, there is no paradox of thrift.

    Where there is something like a paradox of thift, however, is that saving more leads to lower consumption. More than the increase in output from the larger capital stock is required to maintain the capital stock itself. Y is bigger, but C is smaller because I is bigger.

  55. a writes:

    Your description of a process of adjustment seems reasonable. But it becomes a question of how abrupt and deep the planned savings increase is. The greater the intended change in savings, the more difficult it is for the economy to adjust (including adjustment encouraged by lower interest rates). Over time, the economy can shift from the production of consumption goods to investment goods, as you describe. But sudden and widespread increases in attempted saving may overpower the ability of the economy to respond effectively with higher levels of investment. If the firm lowers its prices, it reduces its income (which includes its internally generated savings available to fund new investment). And if the firm further cuts production and employment, a further reduction in income and therefore planned savings will happen for both business and households.

  56. For sure, it is not unequivocal how the shock to consumer preferences will play out. For example, don’t forget that the firm is cutting its price to reduce its inventory, so its revenue, and its ability to finance investment, might be increased depending on whether the price or volume effect dominated. Also, the idea that consumers would continue to attempt to save a desired proportion of their income as their income fell seems unrealistic to me.

    Ideally, economic policymakers would make some kind of cost benefit analysis of whether mitigating pain by braking adjustment was worth the cost of delaying reaching a more efficient equilibrium outcome (assuming that equilibrium is more efficient of course). Even if the economic model that they would use to make this judgement is imperfect, at least it would be objective. Sadly, I suspect that policymakers respond more to the near term, for which they expect to be held responsible!

    Given the uncertainties, I think we have taken our discussion about as far as we can. Or is there more to say?

  57. a writes:

    fini, thx

  58. RYviewpoint writes:

    OK, you’ve found a loophole, thrift as savings can be reinvested. But thrift in the context of a credit crisis means “savings” in an environment where lending freezes up. Fear is rampant, and lending standards are tightened. Less money is available to be put to work because financial institutions are seeking to raise their capital as an act of “battening down the hatches” for the oncoming financial storm. Thus, savings end up in accounts with little or no investment income because it can’t be lent out. Cosequently savers are not rewarded and entrepreneurs have no access to the fear-frozen funds.

    Maybe McCulley needs to rewrite this as the “Paradox of Fear-induced Thrift”. The classic example of fear-induced thrift is putting your money under your mattress.

  59. BigBan writes:

    Oh, Thanks! Really amazing. Greets.

  60. BigBan writes:

    Oh, Thanks! Really interesting. Big ups!