Too much risk?

One of the more depressing bits of emerging conventional wisdom is the notion that the financial system took on “too much risk” in recent years. I think it is equally accurate to suggest that the financial system took on too little risk.

Consider the risks that were not taken during the recent credit and “investment” boom. While hundreds of billions of dollars were poured into new suburbs, very little capital was devoted to the alternative energy sector that is suddenly all the rage. Despite a “global savings glut” and record-breaking levels of “investment” in the United States between 2005 and 2007, capital was withdrawn from a variety of industries deemed “uncompetitive” in large part due to obviously unsustainable capital flows. Very few brave capitalists took the risk of mothballing rather than dismantling factories and maintaining critical human capital through the temporary downspike. Under the two to five year time horizon of our most far-sighted managers, whatever is temporarily unprofitable must be permanently destroyed. To gamble on recovery is far too great a risk.

I don’t pretend to know where all that capital, that incredible swell of human energy and physical resources, ought to have gone. But it doesn’t take an Einstein to know that it probably should not have gone into building Foxboro Court. Sure, hindsight is 20/20. But lack of foresight really wasn’t the problem here. In 2005, how many macroeconomists or big-picture thinkers were arguing that the US economy lacked suburban housing stock of sufficient size and luxury? We gave the building boom the benefit of the doubt because it was a “market outcome”. But the shape of that outcome was more matter of institutional idiosyncrasies than textbook theories of optimal choice. It resulted as much from people shirking risk as it did from people taking big bets.

The big central banks, whose investment largely drove the credit boom, were (and still are) seeking safety, not risk. The banks and SIVs that bought up “super-senior AAA” tranches of CDOs were looking for safe assets, not risky assets. We had a housing boom, rather than a Pez dispenser bubble, because housing collateral is (well, was) the preferred raw material for fabricating safe paper. Investors were never enthusiastic about cul-de-sacs and McMansions. They wanted safe assets, never mind what backed ’em, and mortgages are what Wall Street knew how to lipstick into safe assets. The housing boom was born less from inordinate risk-taking than from the unwillingness of investors to take and bear considered risks. Agencies, asset-backed securities, it was all just AAA paper. It was “safe”, so who cared what it was funding?

Finance is not a closed system, a zoology of exotic contracts and rocket scientist equations. The job of a financial system is to make real-world decisions, “What should we do?” A good investment is a simple answer to that question, with clear consequences for getting it right or wrong. Mom and Pop can have FDIC insured bank accounts, and imagine that there is such thing as a “risk-free return”. But that’s a lie, a sugarcoated subsidy. Foregone consumption does not automatically convert itself into future abundance. People have to make smart decisions about what to do with today’s capital. If they don’t, no amount of regulation or insurance will prevent all those savings accounts from going worthless. When huge institutions treat the financial system like a bank, depositing trillions in generic “safe” instruments and expecting wealth to somehow appear, they are delegating the economic substance of aggregate investment to middlemen in it for the fees, and politicians in it for whatever politicians are in it for. And we are surprised when that doesn’t work out?

Of course we should regulate and manage the risks that were the proximate cause of the credit crisis. Anything too big to fail should be no more leveraged than a teddy bear, and fragile, poorly designed markets should be fixed. But that won’t be enough. We’ve trained a generation of professionals to forget that investing is precisely the art of taking economic risks, then delivering the goods or eating the losses. The exotica of modern finance is fascinating, and I’ve nothing against any acronym that you care to name. But until owners of capital stop hiding behind cleverness and diversification and take responsibility for the resources they steward, finance will remain a shell game, a tournament in evading responsibility for poor outcomes.

Investors’ childlike demand for safety has made the financial world terribly risky. As we rebuild our broken financial system, we must not pretend that risk can be regulated or innovated away. We must demand that investors choose risks and bear consequences. We need more, and more creative, risk-taking, not false promises of safety that taxpayers will inevitably be called upon to keep.


42 Responses to “Too much risk?”

  1. JKH writes:

    You might be assuming too much in one area and too little in another, with respect to the “job description” of the financial system. Most economic investment results from asset allocation strategies of business (including new housing investment, which starts off as a business enterprise, which is then sold as an asset to households). Such strategies are financed either from internally generated funds or externally. But the financing is not the same thing as the investment strategy. The investment strategy is the job of the business manager; financing is the job of the CFO. (Downstream, mortgage financing funds a new house that resulted from some builder’s investment strategy). That’s simplistic perhaps, but it reflects at the micro level a corresponding physical and intellectual separation of the real economic system from the financial system, and a separation of real economic investment from financial asset investment.

    So I would separate the idea of risk taking as between these two planes of the integrated system. Your view that there was not enough risk taking translates, I think, to the idea that there was not enough visionary thinking in the real investment side of the economy. This could include both private and public enterprise. I wouldn’t debate that part of it except to say the counterfactual would seem to involve at lot more central planning in the economy. Otherwise, how can we say that business was responding to anything other than the invisible hand that was dealt to it?

    The financial system in such a world view is quite a separate issue for debate. One of the things that the financial system does is link savings with investment at the macroeconomic level. But this is not the only aspect of its activity. It may not even be the largest aspect. It is involved in several other macro categories of financial activity.

    First is that the financial system connects many economic units with positive saving with those that have negative saving, neither of which has a direct connection to macroeconomic investment. For example, banks offering home equity loans, which borrowers use for consumption purposes, fund them with household unit savings in the form of bank deposits. But neither side of this linkage is necessarily connected to the level of macroeconomic investment. The two sides are brought together as a clearinghouse for the expenditure of current income on consumption of goods and services – not for its net saving and investment. Similarly for the famous mortgage equity withdrawal tsunami, much of which was used for personal consumption expenditures. One might push this analysis to the limit by suggesting that all finance is “inside money”, without any direct real economic effect, and that the true economic transaction is the linking of real investment with ultimate household equity (true saving), regardless of the nature of financial intermediation, but that’s going a bit far with the idea. Nevertheless, the complexity and circuitousness of the financial system reinforces the idea of its separateness from the nature of the real investment it finances. For example, the new house builder and buyer together drive supply and demand for real economic preferences. Finance is the enabler of such a transaction but not the real asset allocator. Teaser rates enabled by Fed policy, and CDOs that manipulated paltry mortgage cash flows are quite distinct from the motivation behind and the result of the granite countertops they financed. Who would question these real preferences other than central planners?

    Second is that the financial system has a great involvement in the trading of financial assets, which has nothing to do with the level of macroeconomic investment. For example, the super-senior CDOs that you refer to I believe are mostly synthetic. As such, they are a bet on the cash flow behaviour of a similar mortgage based financing cash instrument. But like all derivatives, they are a bet involving asset and liability counterparties at the level of the financial system, and again have nothing directly to do with underlying macroeconomic investment. Of course, there are a zillion examples of such derivative bets that are disconnected from real economic investment flows.

    There was a great deal of risk taking in the financial system defined in this way. Most of the ensuing risk malfunction can be attributed prosaically to a failure of imagination in the sense of an overreliance on statistically based risk management systems (i.e. “value at risk” systems), and a corresponding underestimate of capital adequacy. But this failure followed from risk taking that is in a separate category from the heroic visionary risk taking that one might hope for in the real economy.

    A couple of other points:

    I do think one can go a bit far with this idea that leverage is inherently evil. If you really believe that too big to fail institutions should not be leveraged, then you also believe that commercial banks should not be allowed to take deposits. Deposits are liabilities, and constitute leverage to the position of bank equity holders just as much as other forms of debt. Deposits are the dominant source of commercial bank leverage. And of course there’s almost nothing in the way of money supply remaining if banks can’t take deposits. (Is this what Austrians want?)

    Finally, the global savings glut is a croc. This is the “inside” financial system working overtime and globally. China has tons of US assets because it has pegged its currency for trade reasons, and US consumers have responded to such a manipulated pricing basis with a rampage of buying. Apart from those dollars it sells for other currencies, China has no choice but to cycle its surplus back into dollar financial assets. To the degree that China has bought treasuries, its connection with underlying real US investment is very tenuous. And to the degree that it bought agencies, its connection might be just as much with those US consumers who extracted MEW funds through mortgages for the purpose of buying Chinese products, as it is with the amount of US GDP that was actually attributable to excess new housing investment. To the degree that China has done other “riskier” stuff with its money (e.g. Blackstone et al), the relative amount is trivial, and even then is somewhat disconnected from real underlying US investment.

    Thus, the scope of the financial system is in fact larger than the constituent issue of directing saving into investments in the macroeconomic sense of these terms. As such, and with all its failures, the financial system has taken all sorts of risk that doesn’t directly connect with underlying economic investment, and which may have a quite different risk taking problem of its own, quite separate from the heroic failure of real investment risk taking that you have described.

  2. Perceptive comments as usual, JKH. I suspect that the expectations that financial market investors have about returns and risk have been unrealistically optimistic, so that it was difficult to find business projects that measured up. Consumption shifting borrowers, on the other hand, were sufficiently optimistic about their future, or sufficiently desperate, to pay a high enough return to get funded. In my opinion, the central planners – ie the economic authorities – failed in two ways. First, they repeatedly protected financial market investors by cutting interest rates when more extreme risks looked like being realised (eg 1998 LTCM crisis). Second, they could have continued to forcibly divert income into investment by not cutting taxes and spending more on public infrastructure.

    If it is of interest, I have explained these two failures in more detail on my blog:

    (1) bailouts

    (2) government saving

  3. Gigi writes:

    Beautifully said. I have been observing risk aversion in many paces:

    – Management deciding to slightly modifying old products because it is “guaranteed to work” rather than innovating. And this in a high tech industry….

    – Attacking existing markets instead of opening new ones in high tech growth companies. The “market risk” is low and used to justify the program but nobody seems to remember the entrenched competitors. The new markets are ignored as nobody can “guarantee” that it exists.

    – Big pahrma going after slight modifications of existing products to “refresh” their patents rather than investing to discover new badly needed drugs

    – Hollywood’s love affair with sequels, reviving old favorites

  4. David Merkel writes:

    Steve, I think we had two, maybe three things go on here. First, the “originate to sell” model failed because basic underwriting was not done well. The incentives against failure were not left with the originator, i.e., having to hold onto a large equity piece.

    If the underwriting had been done well, the next problem would be weak financing structures on the part of the certificate buyers. Many were leveraged higher than prudent, even on “super seniors.”

    Finally, the servicing models are often flawed. There has to be adequate pay for servicing and special servicing, or else loss mitigation efforts will be poor.

    Risks were taken and avoided, but many of the seemingly avoided risks come back when the one guaranteeing the avoid risk cannot perform.

  5. vince writes:

    JKH, do you have a blog ? A book ?


  6. zanon writes:


    I agree with you that central banks bought AAA securities because they were seeking safety, but why is “I don’t know right now, I want to postpone the decision” an invalid response to your “what should we do?”

    While “a risk free return” may be a sugar coated lie, the demand for a safe store of value is real, rational, and beneficial. Once-upon-a-time, the US government provided this safe store through the good management of the dollar, but that day is gone. While the dollar is still, de facto, the world’s reserve currency, it’s a crappy one, and I can’t blame the world’s savers for casting about for something better.

    Even if you are that rarest of species, an American saver, you’ve had a disastrous decade with no end in sight. Your FDIC insured bank deposits and salary has failed to keep pace with housing, health, education, food, and fuel by dramatic amounts. If man lived by flat screen TVs alone all would be well, but that is not the case. There is no obvious mechanism for you to store value today so you can consume it tomorrow.

    So in a world where nothing is a safe store of value, everything is. You buy some oil to hedge against fuel increases, and some wheat to hedge against food prices. If you don’t own a home, how about a REIT so you’re less short, and maybe you can add some gold just in case it all goes to hell. Toss in a basket of foreign currencies since the greenback is looking sickly. I can’t think of how to hedge yourself against increases in education and health. But note that all of these capital allocations aren’t seeking a positive return, they are just there to avoid negative returns. If the saver could just put their money in a bank account and be done with it, they would, but since Plan A is kaput, it’s time for Plan B.

    I would also add that all this Plan B asset allocation muddies the signal that you expect from capital markets: “what should we do?”. Since the markets cannot respond to “I want to decide later”, they’re convulsing through a “head for cover”


  7. bsetser writes:

    Steve — I would amend your comment to note that lots of investors wanted safety, and a bit of carry. the perception was that the cash flows from real estate lending could be arranged in a way that provided that, while more “speculative” investment on the real side couldn’t, and thus wasn’t as good fit for a leveraged entity like a bank.

    JKH — you are right that Chinese purchases even of riskier assets are a bit disembodied from actual investment. selling an asset in the secondary market isn’t quite the same as new investment. I wish this point was more widely understood. I would though argue that to the extent Chinese purchases of treasuries pulled down us rates, they did encourage borrowing — whether to consume or to invest — and thus were indirectly linked to the increase in investment Steve noted. Or put differently, the fiscal deficits of 03 and 04 and 05 didn’t crowd out housing investment in the way that might have been expected.

  8. It is the nature of investors to attempt to mitigate risk. The whole of financial analysis is attempting to determine where the risk / reward ratio is favorable. We can’t bemoan the investors distaste for risk but we must always point out the folly of anyone ever believing it has been concurred.

  9. Steve:

    I’ve been saying things like this for years.

  10. JKH writes:

    JKH: Good stuff. Do you have a blog or column?

  11. Alessandro writes:

    Steve, excellent article as always! (You are setting the bar very high)

    We have been discussing about bad investment decisions and how those make an economy less productive on Nouriel Roubini blog (by “we” I mean various guests in the comments) and we have reached very similar conclusions. If you look at the aggregate the size of the mal-investment is absolutely terrifying: $500bn of losses at financial institutions on their way to $1 trillion or more due to (with hindsight) obviously bad investments. What could we have done with $1 trillion invested in research?

  12. JKH writes:

    (I did not author the 4:18 comment; not quite sure if the handle error was inadvertent or intended; but no and no)

    vince: thanks for asking, but no and no

  13. Alessandro writes:

    Wildly Off-Topic

    Steve, it just struck me an alternative explanation for the commodity boom.

    You are an unnamed country that suddenly find itself hooked into the greatest Ponzi scheme in history, holding $1tn of dubious debt. You know you cannot hold on for long and desperately need an exit strategy.

    Your exit strategies are:

    1. you stop playing the Ponzi game, the scheme collapses, you and Mr. Ponzi get screwed really, really bad.

    2. you go on playing but require Mr. Ponzi to surrender hard assets going forward (land, companies, etc), you get good stuff from now on and Mr. Ponzi gets screwed. You know you have no chance for this strategy to work, at least not yet.


    3. you go shopping for future production of natural resources (oil, metals, anything) and assume huge dollar denominated liabilities matching your huge dollar denominated assets. You don’t care about the agreed future price, as long as it is in dollars. You ignite a inexplicable commodity boom, but you don’t care, you know you have no other way to get value out of those dollars. Once you are done, you can gently refrain from playing the Ponzi scheme, as long as Mr. Ponzi don’t default on its debt you are ok. Mr. Ponzi and anybody who had priced natural resources in dollars get screwed. You get an amazing 50c on the dollar out of a Ponzi scheme, not too shabby.

  14. JKH — You really should start a blog. I don’t say that to insinuate that your long and thoughtful comments are unwelcome. On the contrary. They are thoughtful and thought well, and you clearly enjoy the conversation. It’d be a public service.

    There’s a tension, when discussing the financial system, between positive and normative aspects. You are right, as a positive matter, that what I ask of the financial system is more than what the professional class standing beneath the banner “Finance” would consider part of their job description. As a normative matter, I think that’s a bad thing. I’m too familiar with what “financial analysts” do, as a profession. Generally speaking, they develop independent opinions tightly constrained by professional norms, in terms of what are reasonable asset classes and assets within classes, that severely limits the possibility that professionally manged capital — that is, most capital — will be deployed outside of what are deemed safe harbors. The “physical and intellectual separation of the real economic system from the financial system, and a separation of real economic investment from financial asset investment” is a bad thing, from my perspective.

    “[T]he counterfactual would seem to involve at lot more central planning in the economy. Otherwise, how can we say that business was responding to anything other than the invisible hand that was dealt to it?”

    I don’t think that’s reasonable. That an outcome occurred in an environment largely free of central planning does not imply that the same outcome would always have occured in similar circumstances absent central planning. The fact that demand for safety translated to a housing boom had everything to do with an institutional environment, with both public and private aspects, that is by no means predetermined. The existence of Fannie and Freddie, and the infrastructure and expertise devoted to generating and securitizing conforming mortgages, served as the seed and template for much of the innovation in the sub-prime, alt-A, and jumbo sector. Housing seemed a good candidate for “instasafety” because it was the only asset conventionally securitized, and had a pseudohistory of “never” having declined on a national basis. It would be quite possible to define tranched paper secured by pooled sector-specific inventories, plant, intellectual property, or subsidiary securities that would have allowed moderate-risk fixed-income investors to make qualitative choices about where their capital would generate the highest margin of safety. Of course none of those would have had the quantitative characteristics to simulate out as AAA paper in rating agency models. But those models were foundationally wrong, and surely not a necessary outcome absent central planning. The scenario we observed was a collaboration between the invisible hand and a set of institutions, habits, conventional assumptions and shortcuts. It was not what “the” invisible hand foreordained.

    That said, the relationship between central planning and free markets in a “good” economy is a challenging one, and I don’t want to accept an implicit assumption that more central planning is always worse. I’m very biased towards market solutions to informationally challenging problems. But there is a “superstructure” (I know! That’s a Marx-word!) that generated constraints and biases under which the “invisible hand” operates, and there’s little evidence to suggest that relying on a meta-invisible hand to get those biases right is better than even very flawed state actors. Of course, I’d be very glad to improve upon flawed state actors, but one must make a case for how, not just hurl epithets about socialism and central planning, to do that.

    When savers finance the current consumption of negative-savers, absent agency costs, they do so only contingent upon assurance that the negative savers will generate suplus future production a portion of which they can skim. That is to say that you are right that current savers are happy to transfer dis-savers future production to current consumption, for a price. And that’s fine, so long as they can discriminate wannabe current consumers with surplus future production from wannabe current consumers without. Rational savers absolutely do not finance the latter party’s consumption, but a lot of savings did in fact go to dissavers without expected surplus production. That was not a matter of volunatry allocation by invisible hands, but uninformed investors and malfunctioning agencies. Depositors would not have made the same loans their banks did. I’m glad to question those outcomes, and againwon’t concede that what flawed institutions have wrought was the only possible outcome absent central planning.

    I agree that there was lots of intrafinancial risk, in terms of leverage and the structure of a variety of assets and derivatives. I framed this piece as “too little risk?” to be provocative. I’d suggest that there was too little risk taken along certain dimensions, and too much risk taken along others. However, even synthetic assets impact the real allocation of capital. If I buy a synthetic bond built out of Treasuries and GM CDS contracts, the teentsy little firm I’m funding is selling protection on GM bonds, driving down the cost of that protection, which affects the credit spreads demanded on “real” bonds, which affects GM’s cost of capital. I won’t make the strong claim that purchasing such a bond is equivalent to actually lending GM money. But broadly and in general, derivative transactions are “passed through” to the real economy by arbitrage and equilibrium, they are not a cul-de-sac unto themselves.

    I’m mostly with the Austrians, by the way, on commercial banks and deposits. There’s this myth that there’d be no investment without highly leverage commercial banks and “maturity transformation” (which is the ultimate juvenile demand for safety that should not in general be accommodated). If all the funds deposited in checking accounts, savings accounts, and CDS were kept as cash in mattresses, that wouldn’t affect the quantity of real capital available for investment. It would undermine the system that we currently have for allocating capital, and the system that we currently use to mediate current exchange. I won’t quarrel that it would be disruptive. But the usual justification for leveraged, fractional reserve banking is that they are necessary to persuade the masses to invest society’s resources, which otherwise would sit unused. That’s BS. As long as wealth is stored as money or future claims, savings does not deplete the capital available for investing. the current banking system has more to do with who decides how capital is allocated than with making capital available.

    I’m not sure if we’re disagreeing at all about the “global savings glut”. Contingent upon certain policy choices, China and the Gulf states have no choice but to invest in USD assets, and they prefer safety to risk with the vast majority. That was my point, it seems to be yours too. They drove up the price of safety, which created a great demand for cheaper (that is higher yielding) safe assets from the private sector, which also largely wanted safety.

    Again, I do agree that the financial system took all sorts of “intrafinancial” risk, and that had much to do with the crisis. That’s why I used the formulation “equally accurate to suggest that the financial system took on too little risk.”

  15. Gigi — Good points all. In my mind, there are a few ways of thinking of all these examples: 1) The opportunities we naifs perceive are not really there, and investment was better allocated to housing and sequels; or 2) Well-managed investment in these “riskier” propositions (without positing superhuman abilities or foresight) could in fact have outperformed (in economic, as well as artistic or moral terms) the actual investments made, signifying some inadequacy in the means by which we collectively make investment decisions. I think (2), but I could be wrong.

  16. David M — I don’t disagree at all. Investors “outside” wanted cheap safety, intermediaries “inside” didn’t have naturally cheap safety, so they manufactured it, with weak underwriting and by leveraging pseudo-risk-free assets into pseudo-risk-free assets. (You can leverage a zero-beta asset all you want! It’s still zero-beta! And perpetual motion machines really work!)

    As I wrote JKH, it wasn’t my intention to claim that “too little risk” was the problem to the exclusion of “too much risk”. The claim is that there was too little risk, by some parties and along some dimensions, and too much risk along other dimensions. (Although the parties taking “too much risk” largely succeeded in isolating themselves individually from the downside of those risks. Those bonuses are not being clawed back. Within the financials, individuals got safety while business entities, taxpayers, and fragmented equityholders got risks they were unaware of taking.)

  17. Zanon — Individually, precautionary storage of a diverse basket of real commodities is a perfectly rational means of hedging the risks to ones accumulated wealth. It is a hedge, not an “investment”, in the sense that the expected real return on stored goods is negative (storage costs, things go bad, accidents and theft happens). One might get lucky if ones basket disproportionately includes some commodity whose relative value dramatically spikes, but the expected return on storing stuff is in the aggregate always negative.

    Which is a problem in the aggregate. The seduction of precautionary storage is a real tragedy of the commons, because we generate greater wealth, activity, happiness, and easter bunnies, if we contrive ways to convert today’s decaying stuff into tomorrow’s new goods, services, and activities. Absent a well functioning financial system, precautionary storage is individually rational but collectively disastrous. That’s why a kick-ass financial system is as important to our collective well-being as reasonable (and not overly fragmented) property rights are in the traditional commons story.

  18. bsetser — I certainly agree that investors wanted safety and carry, but I think we have to distinguish that from the conventional continuous trade-off in risk and return. “Safe” assets were a distinct category of assets (which oddly included even heavily leveraged variants thereof), and were not deemed substitable for “risky” assets, even at equivalent yield. In theory, investors should understand that nondiversifiable risk is priced, and there’s no free lunch. In practice, investors took certain assets to be risk-free (and therefore not worth diversifying, tho’ sometimes they were diversified internally), and looked for the cheapest/highest carry assets within that “safe” category. That makes no sense under any asset pricing model, but I think it’s descriptively accurate.

  19. Ben writes:

    Great article. My interpretation: there’s an unrecognized risk in not thinking for yourself.

  20. MasterPlan (now there’s an ominous name!) — It is of course natural, and not worth bemoaning, that investors want as much safety as possible given for any given level of expected or possible returns. What is worth bemoaning, though, is when investors demand safety at ever increasing levels of return. Perfect safety can’t be had, and conventional safety — no default or nonpayment risk within a stated currency &mdash’ can only be offered at one price, and in quantities determined by the sovereign that prints the currency. When investors demand “equivalent” safety at something other than the sovereign yield, or a greater quantity of “risk-free” assets than the sovereign is willing to offer, that is quite unreasonable, and merits a bit of bemoaning and belly-achin’, no?

    IA — I’m not at all surprised (and certainly glad) to hear that…

    Alessandro — I think the calculation you’re making is one that’s been made. There can be all kinds of ancillary benefits from getting taken by a Ponzi scheme, and depending on where one gets in on such a scheme, one doesn’t have to lose. To bring it back to the topic of this post, some of the investors seeking “safety” were really not enthusiastic about their investment, not only indifferent to what they were funding, they were not so much interested in achieving any return at all, but merely limiting and rendering predictable their loss. Given the ancillary benefits that generated these negative-return investments, these investors’ participation may well have been rational. But from the perspective of the economy that received the capital, the indifference with which it was allocated might have made a negative contribution to the overall efficiency of the capital allocation decision.

  21. This might be a shocking idea to Americans, but saving with a negative return is not necessarily irrational. It is rational to substitute consumption inter-temporally to optimise your welfare, and it might be worth paying to defer consumption if you expect your needs to rise, or your income to fall, a lot in the future. It is true that currency sets a zero floor for return, but that is a nominal zero, not a real zero. So it might be quite reasonable to stockpile commodities for example if you cannot trust anything else. The example I always like to give is a squirrel – when a squirrel buries nuts, it is not doing so with the intention of growing nut trees!

  22. Alessandro writes:

    Steve, London Banker over at RGE has a post on how we reached here, what to expect for the future and what it needs to be done. His history matches most of your points and his gloomy future scenario looks all to probable to me.

  23. JKH writes:

    SRW – thanks for your kind words and response. Your “normative” perspective makes sense. To the extent that the separation of finance from the real economy results in a bloated, redundantly complex, ineffective financial system, it is not a good thing. I agree with your point that the derivative space affects the cost of capital for “real finance” via arbitrage. And your point about China et al driving up the “price of safety” is right on. Finally, I also agree with your point, “I don’t want to accept an implicit assumption that more central planning is always worse”. My use of the terms “invisible hand” and “central planning” was a bit overcharged; the intent was more nuanced. I’m all for reasonable constraints and law and order; e.g. intelligent, coherent, regulatory capital adequacy rules for the entire financial system.

    One area of difference remains regarding current interpretation. If the housing bubble was a concentration of real economic demand, then misguided finance facilitated it, pumping up both real and financial bubbles. Dilution of such macro concentration risk through alternative investments as you mentioned would have constituted sensible diversification of real economy output and investment. Consistent with the normal benefits of diversification, it would have reduced risk in both the real economy and the financial system. But I think my point is that the real economy seemed to have trouble identifying these opportunities on its own. Finance usually responds when business has sufficient desire to capitalize on such ideas. Would it have been difficult to finance such opportunities had they been actively identified by real side entrepreneurs and businesses? I’m not so sure. Clearly, the financial system failed miserably and compounded the bubble trouble in housing. But I’m not sure how it failed in terms of a lack of financing for alternative real investments. The type of diversified investment and financial application to which you alluded should appeal to prudent risk adjusted financial demand. But what happens when there is a drought in the supply of such desirable paper simply because there is not the requisite desire for real investment by potential issuers (compared to a surging, hyperbolic lust for granite countertops?) To be judged normatively bad, this would imply in my view some sort of public failure to right the ship by stimulating such alternative investment demand (whether or not such intervention is desirable).

    Your Austrian point was most interesting. I’m familiar with the area only at the edges (hence my reference to it as a question). I sometimes wonder what an “Austrian conversion day” for the banking system would look like. As a subset of this question, the view has always intrigued me that the Federal Reserve’s control over the Fed funds rate should be ceded to the markets, along with the stronger version that the Fed itself should be abolished. James Grant is one of the more eloquent and vigorous proponents of such a view. But I’m suspicious of the idea. The puzzling aspect to me is how the result would actually work, and what the full implications would be. I’ve never seen it fleshed out as an operational model. (The idea tends to come across as a loose prescription flowing from rants about the inability of Greenspan and Bernanke to have forecasted economic activity accurately, or to have set the Fed funds rate at the “right” level.)

    E.g. how would the big clearing banks like Citigroup, JP Morgan, and Bank of America effect the payments due each other as a result of the payments due to their depositing customers? How would they handle intraday credit risk on such payments? What would happen to the base money that is now used to make such payments? Does the state become the clearing system? If so, how is the aggregate supply of balances required to make interbank clearing payments determined? What is the effect of such supply, however determined, on short term interest rate volatility, which is now left entirely to market forces? Does conversion invoke the complete abolition of commercial banks? Or does it suggest the state confiscation of 0 maturity money balances only, leaving commercial banking to operate on a strict matched maturity basis?

    (As an aside, your imagined mattress saving system could become the bridge between saving and investment, but if the hoarded paper is issued by the central bank/state, the state becomes the financial system. It would seem that some maturity transformation risk would now reside in the state’s operations as the designated macro financial intermediary.)

    The above machine gunning of confused, thorny details and questions is as much for the record as for your immediate consideration. One has to be in the right mood to even begin to think about it (unless by chance you’ve already pondered a hypothetical architecture for such “conversion”).

  24. I do not think that China bid up the relative price of safety. If China had been the driver of the process, their influence would have been transmitted through wider spreads during the boom, not narrower spreads. I explain the argument in more detail in a post on my blog.

    An explanation that fits the facts better is that the origin of the process was the boom in spread product driven by financial innovation and complacent investors, which fed a consumption boom that sucked in Chinese imports, which, given their peg to the dollar, led the Chinese to buy dollars and then reserve assets.

  25. Benign Brodwicz writes:

    In a world of fiat money gone mad (and bad), you have to include a “bubble premium” in your discount rate on all “investments,” and mind your trailing stops. Bubbles will always suck funds away from more roundabout, long-term real capital formation that produces a real return because people almost always fail to discount the bubble correctly.

    We are living through the biggest “investment” mania in history. Last time around it took a depression to get people to swear off speculation.

  26. groucho writes:

    Steve, excellent ass kicker(as usual)

    Too much…risk or too much moral hazard?

    Too much…risk or too much “gaming Globalization”?

    Too much…risk or too much Ponzification?

    Too much…risk or too much P.T. Barnumization?sucker born

    Too much…risk or too much special interest?

    Too much…risk or too much Quid Pro Quo? state finance/CB

    Too much…risk or too much CB back-stopping bad finance?

    Too much…risk or too much i-rate manipulation?

    Too much…risk or too much dollar hegemony?

    Too much…risk or too much printing press?

    Too much…risk or too much DISHONESTY?

  27. Mises writes:

    Steve, I must ask if you’re familiar with austrian economics and or their business cycle theory? your last few posts have been very indicative of at least a very similar structural understanding.

    if not i suggest you read over this

  28. Henry writes:

    Crowded trade, the quest for “safety” became.

    Crowded trades are expensive to buy into, and easy to lose capital in.

    When everyone goes looking for (especially when arbitraged by fee-taking young boiler-room dudes in cheap suits driving Ferraris) “safety”, time to run the other way.

    Key to the real estate bust was the transformation of Capital instantly into Consumption, with consumer homes mis-labeled as capital investments.

    That John Stuart Mill quote about “misallocation of capital”; the capital is already lost. It was lost the day it was loaned to CONSUMERS to buy a CONSUMPTION item that forseeably could not hold relative value throughout a 30-year span of aging occupants, declining production and a vanishing Empire.

    Now they’re just trying to allocate the “haircuts” and who ponies up the lost capital into soothing various “black-hole” institutions with the political clout to get labeled TBTF.

    There’s been only one investment to hold since 2001.

    And there have been very few businesses into which intelligently invest capital since then.

    We are in the process of rebooting our monetary system, called “Dollar” or other, with any Fed/IMF balance sheet gold actually still in hand, as when the world left the Pound behind in 1930s for the dollar. But the monetary transition striptease will require employing every other fan or veil before getting down to that final revealing.

    Question personally is, do YOU hold the capital that will be in demand when the moment of such transition is upon us? Then you get to be part of the system reboot, rather than an old floppy drive getting thrown away…

  29. RYviewpoint writes:

    Isn’t “risk” a concept like “beauty”, i.e. it is in the eye of the beholder?

    OK, I’m jesting, but my point is serous: Any attempt to define risk comes down to identifying a model that allows you to evaluate “the risk”.

    But any model that purports to quantify risk is in fact a construct that embodies assumptions about “relevant” factors. I.e. factors abstracted from the current situation. These factors then must be evaluated against previously abstracted events drawn from historical experience and expressed in a theoretical model. But how do you validate such a model? And how do your justify the selection of the relevant factors to model?

    Perhaps if you have a long historical series of events that can be shown to “fit the model”, then you will gain confidence that you have found the right factors and modeled them correctly.

    But isn’t this just a fallacy of induction? (Our long experience of sunrises and our model of an earth rotating on an axis gives us confidence that tomorrow there will be a sunrise. But what if a black hole swallows us tonight?) What assurance do I really have that the model correctly abstracts all relevent facts (we can ignore black holes)? What confidence do I have that my series of interpreted events (my belief that the earth is rotating and a sunrise represents my patch of earth rotating about to yet again to face the sun)? What justifies me in using the model to step beyond previously evaluated events to characterize this current event (to claim that the sun will rise tomorrow)?

    How do you validate a model of risk. It is easy to give a “mathematical validation”, i.e. if you assume the model then you need simply focus on demonstrating the correctness of your calculation of current events from past events using the model. But that is self-contained mumbo-jumbo. The real issue is: How do you validate that the hypothesized “model” in fact models the real world. How do we know that our risk model is the right model for the current situation?

    Any talk of “too much risk” or “too little risk” is a stretch. All such arguments presuppose that we have settled the issues of (1) how to characterize the relevant facts and (2) which model validly represents the current situation. But these presuppositions end up looking very much like arguments about whether something is “beautiful” or not.

  30. All — Apologies are cheap, since I don’t change my ways, but sorry as usual for disappearing.

    Ben (way above) — I think you’re absolutely right.

    JKH — I don’t disagree with you at all that whatever “real economy” issuers were out there with “sellable” projects did get funded during the credit boom. Loose money was loose everywhere. As you say, the problem had to do with the mix of projects out there that were sellable, that is, that an entrepreneur proposed in some manner that a “prudent person” manager could reasonably put other peoples’ money into. You’re quite right to point out the aggregate diversification issue: any reviewer of US aggregate investment during the boom would be nervous from a portfolio-risk management perspective. Almost everything was correlated to real-estate one way or another. If we’re going to draw boundaries for blame, one might then argue that we should blame an absence of good entrepreneurs for offering the financial system too few ideas to invest in. But I think that’s letting the finance side off the hook much too easily.

    As we saw in the 1990s (to our peril, but to our relatively mild peril), when there is money that wants to invest in a sector, entrepreneurs pop out of the woodwork to take it. What the financials ought to have been doing these past few years, what is frankly their job in the universe, is to “think big” about long-term aggregate investment needs and start up venture cap funds, incubators, recruiters, all the machinery that diminishes the chasm of entrepreneurship, in sectors they knew had long-term potential. They know how to do this, but it is an intrafinancial problem that delegated money is uncomfortable doing what hasn’t already been legitimated by the fact that others are doing it. That money seeks the safe harbor of the herd — and crowds that harbor until there is a catastrpohic collapse — is an outgrowth of the degree to which we have professionalized the investment process, and mix of incentives and hazards we have created for managers.

    On the Austrian-ish stuff, I’m not a big believer in “flag days” (a term I owe to Mencius Moldbug). I do think fixing the financial system — that is building a system whose asset prices more faithfully respect the real economic value of what backs them and which better directs real economic decision-making — does require foundational changes in the structure of banking and money. But I don’t think we get there in one fell swoop, I think we evolve to it. I think we have to introduce a variety of alternative practices, new media of exchange, “point-of-sale” technology that permits sellers and buyers to price and pay in different “currencies” conveniently, infrastructure for managing assets that represent claims on future goods and services rather than claims on future currency as stores of value and hedges on future consumption, etc. We don’t have to eliminate existing banks or the Fed or whatever, but we do have to curtail their ability to define the terms of commerce in ways that prevent alternatives from outcompeting them.

  31. Rebel — I think it’s too bold to say that any one thing was the driver of the “safety” bubble, but I don’t buy your argument for exonerating foreign central bank purchases. I do agree, broadly, that their policy choices were the outgrowth of currency pegs made and maintained for reasons other than affecting dollar asset prices, but, regardless of intentions, they had a large effect. Re the specifics your post, I think that the proxy for “spread” you use isn’t great. Swap spreads say something about the borrowing costs of large banks relative to the borrowing costs of the Treasury, but they’re not a good proxy for nonfinancial risk. The banking system was considered “safe” to lend to (a view we are validating ex post), and it’s not at all clear that central bank demand for a subset of safe assets would push up the prices of that subset more quickly than it would push up the prices of close substitutes.

    Here’s an analogy. An ice cream seller offers chocolate and vanilla. Consumers value their first scoop of ice cream at $4 and a second scoop at $2, and will pay a $1 premium for chocolate.

    One day, there are 5 customers, 5 chocolate scoops, and 5 vanilla scoops. The vendor sells the chocolate scoops for $5, and the vanilla scoops for $2.

    The next day, along with the original customers, there are 5 new customers who will only buy chocolate scoops, but are willing to pay $6 for them. The vendor sells all 5 chocolate scoops to the new customers, and sells the remaining vanilla scoops to the old customers, now for $4 each.

    The new demand was solely for the “premium” chocolate ice cream. But the price of chocolate ice cream only changed from $5 to $6 (20% increase), while the price of vanilla ice cream doubled (100% increase).

  32. Benign — Implicit in the caveat implied by your use of the term “fiat”, it seems like you think that bubble investing is not an irreducible fact of human psychology, but might be amenable to a better monetary / financial system.

    I don’t disagree at all that the one we have seems to consistently generate bubbles. Most defenders of that status quo would say that it’s utopian to hope to eliminate (or even much mitigate) financial bubbliciousness. After all, tulip bulbs happened under a gold standard. I’m more optimistic than that (though I don’t think that “sound money” would be enough).

    Do you think there’s any way out of the fact that talented con-men are capable of making worthless paper that is difficult to distinguish from claims against genuinely valuable future projects? Is it possible to devise financial markets that are efficiently foresighted, or must we resort to greater central planning (which has its own weaknesses)? Are we simply incapable of evaluating and funding roundabout projects without tolerating a great deal of expropriation (either by corrupt public officials or by artful thieves)?

  33. groucho — yeah… too much of lots of things…. and too little of lots of others. “risk” is too broad a word to make a quantitative distinction.

    Mises — I think I do tend to be pretty “Austrian-ish” in my thinking, although I know there are specific details I would quibble over, and I’m more open to certain kinds of government activism than most people who self-identify as Austrian. That said, my view of what “Austrian” means has developed haphazardly, and I’m delighted by the reference, which looks like a nice way to go a bit deeper. Thanks!

  34. Henry — I think the “one investment to hold” became a crowded trade too recently, and it hasn’t been so easy to hold (though I do hold it, if I’ve guessed right). I love “the monetary transition striptease will require employing every other fan or veil before getting down to that final revealing.” I think that’s almost just right — we are dealing with our problems with a combination of hopeful denial and cynical cost-shifting that amounts to a bewildering dance likely to leave all but the most clever or cautious wounded. I don’t think there will ever be a final revealing… if the one asset to hold were ever to be so plainly vindicated as such, it would become liable to confiscation, one way or another, and the dance would continue.

    After all, the point of the dance is to perform a reboot without anyone noticing, and to ensure that the costs and gains seem fair, like acts of nature, even though they are not acts of nature and can never be fair.

  35. RYviewpoint — That’s exactly right, and interestingly put. The point of this piece wasn’t as much to claim that there was “too little risk”, but to point out that risk is a multidimensional phenomenon, and that policy changes intended to “limit risk” are as likely to do harm as to do good. We need to get out of the habit of imagining that complex phenomena can be reduced to a number, that a good thermostat would solve all the world’s problems.

  36. Benign Brodwicz writes:

    Steve –

    A hard money system is indeed subject to bubbles. But I would submit that a “democratically controlled” fiat system has generated more bubbles than any hard money system.

    It’s like the problem we have in Congress: tax cuts sell, spending cuts don’t; inflation that makes people feel rich sells, monetary or credit discipline that risks being blamed for a recession and possibly causing the loss of the Fed’s illusory “independence” doesn’t.

    Whether in Congress or the Executive or the Fed, we’re being “governed” by short-run political capital maximizers (and the judiciary is hardly non-political, but they don’t have their fingers on the purse strings).

    A fiat money system just makes it too easy to screw up. Anyone have a reference on the prevalence of bubbles in fiat vs. hard money regimes?

  37. JKH writes:



    You’ve written something that would be a fine template for the response of any corporate risk committee to any first draft risk management presentation.

    It would have been particularly good fodder for the PHDs at LTCM, as well as for the planetary hordes that feasted on CDOs.

    It’s the past event extrapolation and associated facile construction of standard deviation that was the seductive killer in both cases.

  38. Steve,

    I suppose that it would be possible to construct sufficiently weird consumer preferences and supply response to get spread narrowing on additional central bank buying of premium product, but I find your example particularly implausible. Your non-central bank consumers seem to ignore their $1 price premium for chocolate when they take their first scoop, such that they end up paying a $3 premium!

    What would you regard as representative of spread product? If data for it is readily available, I would be glad to repeat my analysis using that instead of swaps.

  39. JKH writes:

    SRW/RE –

    I interpreted Steve’s point in the context of the absolute price of safety (i.e. US Treasuries) being big up in comparison to a counterfactual that might have excluded such large concentrated purchases of the same. I think RE you’re making a wider point about the cross sectional spread between Treasuries and risk product (including “near-Treasuries” such as agencies), while also reasonably questioning cause and effect in the process.

    I used to hammer away at the global savings glut issue on Brad Setser’s blog. (I ceased more or less because both his patience and knowledge of the issue exceeded my persistence.) Not to get into the detail of that rant here, one of my suspicions was that the effect of foreign central bank purchases on US Treasury yields was generally overestimated. I must admit I’ve not read the published material on this issue, some of which I gather uses regression analyses to estimate what the “normal” Treasury yield should have been, absent such unusual central bank intervention. My own opinion is only intuitive. But what does make more sense to me is the slightly modified argument that the concentration of foreign central bank purchases in a particular asset category would have an effect. The directional conclusion may be similar, but the argument is nuanced differently. The usual story I think is that massive capital inflows per se pushed down interest rates. My interpretation would be that concentration of such flows in a particular asset category such as Treasuries would be the more precise explanation and necessary condition for lowering yields relative to some more “normal” counterfactual. By way of contrast, it would be interesting to consider another counterfactual wherein the same global imbalances and foreign surpluses were invested according to a more balanced asset allocation strategy, including risky debt and equities; i.e. in reasonable percentage proportions across different asset classes for the trillions to be invested. I wonder if in such an imagined system of foreign official asset allocation, the interpretation of the capital inflow effect on US interest rates in general would have been quite the same. Indeed, the allocation of foreign official flows is now proceeding slowly with such an expansion along the risk curve, but at a relative snail’s pace, as I think Brad might confirm.

    For example, a critical characteristic of the huge Chinese capital flow into the US has been its relative concentration in Treasuries and agencies. This created a problem in terms of risk diversification for US domestic players. China fell far short of any reasonable benchmark for global investment diversification, and was far too concentrated in the iconic US “risk-free” asset. This created a financial engineering demand for the manufacture and sale within the US of “near risk-free” fixed income product, such as was attempted in the construction of senior CDO tranches. This imbalance between foreign concentration and the knock-on domestic search for synthetic diversification would seem a likely contributor to identified distortions in spread markets. It is still possible in my view that Treasury yields were lower for the reason I indicated, and that risk spreads also narrowed due to a desperate search for near Treasury quality in combination with the financial system response to that search. But the critical driving force was the anchoring of the most extreme imbalance (i.e. the “savings glut”) in the risk free benchmark for US interest rates. And given the importance of dollar capital flows and the dollar bloc generally, this arguably could be interpreted as an even broader distortion in the implied risk free benchmark for global interest rates.

    I don’t see spread product as having driven the housing boom per se. The housing boom was financed in the first order by mortgages, including sub-prime mortgages. The tiering of this first order mortgage risk into low and high risk cash flow tranches fed a demand to fill a vacuum of low risk product created by foreign central bank absorption of the same. Otherwise, under conditions of more balanced asset allocation as in the counterfactual noted above, regular unreconstructed mortgage financing might have fed the same type of housing boom.

  40. Rebel — My conjecture is that “safe” asset buyers traditionally had expectations regarding absolute return, that a combination of low interest rates on the short end and CB buying on the long violated. That is, low-risk investors were presented with the choice circa 2004 of either recalibrating return expectations upon which a great deal of future planning was based, or to become “risk” investors. That combination created demand for something that was not “risk”, but that also was not Treasuries. A class of assets for which there was little demand (low coupon, but not Full-Faith-And-Credit or AAA corporate) suddenly had a new clientele, and its price moved sharply, more sharply even than the Treasury prices whose up-bidding was the catalyst for the change.

    My ice cream example is very contrived, but I don’t think it’s as bad as you think. What you’re objecting to is price discrimination (in sequence rather than over consumers here), that is, I’m presuming the supplier gets the full consumer surplus on both scoops. Two scoops of ice-cream are worth $6 to our consumers, and chocolate is worth a buck more on any scoop, and that’s exactly what our supplier gets. Given the a fixed supply of both commodities and a strategic monopolist who knows consumers’ preferences, this should be achievable.

    Your objection that I’ve really posited a $3 premium for chocolate implies consumers could have opted for a greater quantity of the inferior vanilla, which (by assumption) they can’t. If we let the quantity of vanilla be unlimited and set to a uniform price in a competitive market, then sure, the price of vanilla would be its marginal cost, and the price of chocolate could be no more than $1 + the price of vanilla. But with fixed supplies and a monopolist, consumers would buy both scoops if offered at $4.99 for chocolate and $1.99 for vanilla, because the cost of only buying the cheap vanilla is a forgone $5 in potential joy for a savings of only $4.99. (The total achievable joy is two scoops, one of ’em chocolate, worth $7, and eating vanilla only gives ‘en $2 worth.)

    In the real case, Treasuries and “safe” private paper, the supply of Treasuries is (in the short term) fixed with respect to market demand, while supply of the substitute is potentially unlimited, but clearly subject to a lag. (Remember, Wall Street worked desperately to increase the supply of this stuff — it was obviously profitable to make, and supply did not quickly expand until putting together deals was no longer profitable.)

    Let’s try again. There are three kinds of consumers. Everyone likes chocolate at least as much as vanilla, but [Type 1] values chocolate at $6 and vanilla at 0, [Type 2] values chocolate at $5 and vanilla at $4, and [Type 3] values both types at $2. Health-conscious consumers value a second scoop at zero.

    Initially there are 5 scoops of each available, 6 consumers of [Type 2] and 10 of [Type 3]. Suppliers know these preferences, and are (short-run) monopolists. Their revenue-maximizing menu is $4.99 chocolate and $2 vanilla ($25 + $10 = $35).

    Next round, everything is the same, except two [Type 1]s enter the market, so there are 2 T1, 6 T2, and 10 T3. Now the revenue-maximizing menu is $4.99 chocolate and $4 vanilla, [$25 + $12 = $37]. The price of chocolate has not changed at all, but the price of vanilla has doubled, even though all the “new demand” was for chocolate only.

    I don’t think there’s anything that weird or strained about these stories. We need to posit market power and clienteles with distinct preferences, both of which seemed to exist in credit markets. (CBs and private safety-seekers behaved differently, Wall Street “engineers” strategically bundled and priced deals in a manner that gave them economic profit.)

    I can’t tell you what a better measure would be because I don’t think the test works, at least not without making some explicit assumptions about the elasticities and cross-elasticities between Treasuries and other safe assets, which implicitly means assumptions about the clienteles buying and selling these assets.

    I feel (I often feel) like I’m being an asshole, because you’re trying to test something, and I’m ignoring the result and saying it can’t so easily be tested. I think it plausible that CB treasury buying might have raised the price of Treasuries more than that of other safe assets or less than that of other safe assets. I think if you were to look, you’d find that in relative terms, yields compressed more for “safe” assets as a category than for “risk assets” (e.g. low investment grade/high-junk bonds) over the period of CB enthusiasm, so the spread between should have widened. But even if so, a quick glance doesn’t unambiguously bear that out, it looks like business cycle considerations overwhelm whatever impact a monotonic increase in CB buying might have had on the data. (I should be using the 30yr Treasury series, but for some reason I can’t make it work passed a few years ago.)

    I do take the “conundrum” as evidence if CB buying, exactly the conjecture you are trying to refute. Maybe the thing to look at is term-premium on non-AAA corporates vs term premium on Treasuries, on the theory that long corporates would have widened more with increasing short interest rates than long Treasuries, as CBs play (though not exclusively) in longer Treasuries but not longer corporates, and non-AAA corporates and treasuries are less direct substitutes than Treasuries and other “safe” assets. But honestly, it’s much easier to poke holes in other peoples’ tests than to come up with a good one. I apologize for taking the easier and more annoying side of the debate!

  41. Steve,

    You are assuming that the supplier has monopoly power and that the consumers do not trade between themselves, which as a representation of the US bond market seems a bit far fetched to me. My preference is to look for the most plausible explanation, rather than what is possible – as Einstein said, “everything should be made as simple as possible, but no simpler”. For what it is worth, I prefer to think in terms of an Edgeworth box. When a consumer enters with very strong preferences for safe bonds, they alter the equilibrium marginal rate of substitution between safe and risky bonds.

    Of course we can shoot holes in each other’s arguments, but in this case, I think the burden of proof should rest first with complex arguments that yield a conclusion that foreigners are significantly to blame for the credit crisis in the US (and the UK, which, by economic culture, is practically the 51st state).

  42. Rebel — I don’t forbid trade between consumers, I only conjecture limited supply and specific preferences among different clienteles.

    I agree very much with the Einstein quote, it’s one of my faves actually, but we’re focusing on different halves. You say, given the price behavior you’ve observed, parsimony suggests we reject the CB-demand hypothesis. I say, given other aspects of the price behavior we’ve observed, your argument violates the “no simpler” clause. I cannot find explanations I find remotely compelling for the disappearance of bond vigilanteism during this decade other than CB demand, and I view implausible the conjecture that the extraordinary capital-flows they’ve provoked have been entirely inframarginal and irrelevant to the unusual price behavior we’ve observed.

    I think we’ll have to agree to disagree here. I’ll concede a tip o’ the hat, because you tried at least to take data to a model (and I certainly have not), but I won’t alas concede that your model lives up to the “no simpler” rule.