Wednesday, July 5, 2017

Mallaby, the Fed, and technocratic illusions.

One of the frustrations -- or perhaps challenges -- of studying monetary economics and monetary policy is howFed talk and writing on economic mechanisms, causal channels, and effects of policies is far ahead of our actual, scientific knowledge. And writers outside the Fed go leaps and bounds beyond the Fed in advocating strong policies based on the latest stories.

A good example is Sebastian Mallaby, author of "The Man Who Knew: The Life & Times of Alan Greenspan," who wrote last week in the Wall Street Journal Review, that the Fed should surprise us more.

His basic idea: the Fed should monitor asset prices; diagnose when a boom turns in to a bubble; and then actively suppress higher stock prices. And, in addition to interest rates, asset sales, "macro-prudential" regulation (telling banks to stop lending), the Fed should deliberately surprise markets more, adding volatility, in place of central banks' and governments' centuries-old quest (often illusory) to smooth asset prices.
By being less transparent—and reserving the option of deliberately ambushing investors with a shock move—the Fed could discourage them from taking too much risk. 
The painfully learned lesson from the late 1990s and mid-2000s is that excess financial serenity leads to excess risk-taking, which in turn increases the chances of a blowup.
But the equally hard lesson of 2008 hasn’t yet been absorbed: that they [the Fed] should embrace modest, short-term market instability to head off truly disruptive crashes over the horizon. Instead, the calmer markets remain, the prouder the central bankers feel.
Mr. Greenspan and his colleagues faced the danger that the interest rate that would stabilize consumer prices would also destabilize asset prices. The Fed could have escaped this dilemma by acting less predictably. Instead, it telegraphed its intentions and avoided surprises.
Rather breathtaking, no? The last paragraph adds more -- when the Fed wants to lower interest rates to stoke the economy, that causes "bubbles," and the Fed should offset the bubble with deliberate volatility. Hit the gas and the brake at the same time. Greenspan wasn't obscure enough.

What's wrong with "bubbles" anyway? There is one sensible comment,
..when risks seem modest, Wall Street borrows to make bets that look great based on the Sharpe ratio.
Financial crises are always and everywhere about debt. But if Wall Street debt is the problem, just what is the entire Dodd-Frank apparatus to monitor Wall Street debt all about? Really, if Wall Street defaults are the problem, is deliberately inducing volatility to your and my portfolio the answer? Would not a little more capital be a better idea?

Academics do not know exactly how the financial system works. What I as an academic do know, a little more than the average person, is the limits of knowledge - just how much is not known, what the holes are in stories bandied about, and which stories have no basis yet in theory, experience, or evidence. An academic knows that many stories about how the world works are wrong, and we know that many other stories might be possible but have not been written down coherently and evaluated against experience. Knowing what you don't know is knowledge.

It is amazing in that context how much people advocate strong public policy actions -- actions that cost a lot of money, and threaten to put a lot of people in jail -- on stories that are either demonstrably false, or as in this case have no scientific foundation beyond cocktail party speculation, and many glaring logical holes.

For example, it is commonly bandied about, as in this article, that low interest rates induce investors to "reach for yield,'' and create "bubbles'' in asset markets. This is stated as a known, scientific, fact. I know, though it may be true, that this is not yet a known fact. Known facts have to start with a mechanism. Just what is the mechanism? Borrowing at 1% and lending at 3% is exactly the same as borrowing at 5% and lending at 7%. What connection is there between the level of short-term interest rates and the risk premium reflected in the differences between prospective rates of return on different assets?

Well, there are stories about it. Many theory papers have done so in the wake of such speculation. It takes a lot of friction carpentry -- only leveraged intermediaries hold assets, and a lot of nominal illusion or accounting constraints, so that 7-5 is not equal to 3-1. Yes, past booms have involved credit in some way. But most of the time low interest rates correlate with busts, not booms. There are empirical papers,  that seem to show some effects, with all the caveats about empirical work in economics. But none of this elevates it to a known and verified fact, ready for exploitation by policy makers. [I foresee also a swarm of comments opining that yes, low interest rates cause asset booms, thereby missing the point that we shouldn't make policy on such opinion, but rather on well understood causal channels.]

Good policy waits for some sort of scientific evidence. We don't want the government jumping on every food fashion that comes out of the organic farmer's markets of Palo Alto either.

And if the idea that the Fed has the technocratic competence to understand the difference between "boom" and "bubble," the political mandate to determine the correct level of stock prices -- something that affects many voter's pocketbooks! -- and is ready to exactly offset its manipulation of  short term rate by deliberately injecting just enough volatility to hold down prices... Well, I titled the post "technocratic illusions" for a reason.


  1. The whole concept of "normalization" seems to be based on a technocratic illusion.

  2. It might be the Fed is really powerless to alter the unconscious urge to herd in financial markets, powerless to mitigate competition to out-lend the other guy, and powerless to restrain the irrational optimistic outlook of humans. Hence boom and bust might be the natural economic tendency.

  3. I largely agree with the content of the post.

    However, I would quibble with the paragraph on reaching for yield. I never thought that argument was about leverage. Rather, I thought it was about pension funds that have return targets. If they are promising to pay out 5% in the future, but can only earn 3% in investment grade corporates, then they might need to take more risk to get the 5%.

    That being said, I'm not so sure on the connection between this portfolio shift and bubbles. The traditional story is that bringing interest rates below the natural rate would lead to an investment boom. That boom is a temporary overinvestment compared to a world without the stimulus, not necessarily a bubble. Bubbles are not booms. Bubbles are more about overvaluation.

    1. You are right. I didn't want to take the post off track into speculating about whether the low interest rate causes bubbles story is true and if so how. But the idea that pension funds and endowments have fixed nominal rate of return targets, and lever up to achieve them is often offered. Of course that makes no economic sense either, so it rests on deep nominal illusions of a mass of investors large enough to move prices. Which may also be true, but is also not understood well enough for the Fed to include in a technocratic model so finely calibrated that the Fed should ratchet up and down deliberate volatility to offset it. Interestingly, by focusing on Sharpe ratios, Mallaby denies this mechanism. Fixed rate of return investors ignore volatility. That's another problem of cocktail party stories. They contradict each other, but it's easy not to notice the fact.

    2. I am an avid reader of this blog and normally find myself in violent agreement with its perspectives, but this time I must part company.

      No, I am not saying that I agree with deliberately introducing more volatility into the markets as a way of avoiding bubbles. But, having spent most of my professional career either on a trading floor or otherwise working with institutional investors, I can assure you that the "reach for yield" as a cause of asset bubbles is a real thing and it was, I would argue, the base driver of the bubbles that proceeded the GFC. You may argue that this is irrational, but although I was also deeply schooled in the EMH, I cannot possibly justify the price of Tesla shares either (which is why I am short them since I think that eventually the market gets it right, although shorter term deviations certainly appear to be possible).

      If you are looking for empirical evidence of the "reaching for yield" phenomenon, I think that it would be interesting to study portfolio compositions over interest rate cycles. I suspect that, at both the institutional and retail levels, portfolio allocation systematically becomes riskier when interest rates are low AND the economy is still growing -- that is, we need to exclude environments of low interest rates due to poor economic performance producing risk aversion. This would be, I believe, evidence of a "reach for yield" shift caused by financial repression of the type we are currently experiencing and that was a factor pre-GFC.

  4. Borrowers to depositors at the Fed balance sheet is out of wack with one borrower and thirty depositors.

    If you believe interest flow and pricing are related, then statistically the variation in interest charges only covers half the pricing possibilities, mainly the depositors are reacting to the price of government goods as government is the sole borrower.

    If government repricing occurs as slowly as the CBO predicts, then we are talking about waiting 50 years for government to get is borrowing in order, 50 years of very little price discovery in the private sector. I say 50 years because that is how long its been to reach this state since the last monetary regime change in 1972.

  5. Wow. What a great example of this pernicious thinking. Finance really is still at medieval levels of understanding.

    Recently, many people noted that Tesla had a larger market capitalization than Ford. I was pretty amazed by that, so I looked up their financials.

    Ford has a market cap of $46 billion plus $143 billion in debt. Tesla has a market cap of $58 billion and $7 billion in debt. In other words, Ford is 3 times the size of Tesla ($189 billion vs. $65 billion), but claims on their assets are mostly in the form of debt instead of equity.

    Now, do you suppose risk-seeking investors choose to invest in Ford over Tesla because they like how the high level of leverage gives them higher returns even though that leverage is dangerous?

    Do you think a risk-seeking, over-optimistic market would have more Fords or more Teslas? And, thus, do you think a risk-seeking, over-optimistic market would have more debt or less debt? Would it have more equity or less equity?

    Investors in Ford are mostly seeking a safe, certain cash flow. They see some big giant buildings with expensive equipment and they figure that, even if Ford doesn't make a profit for its shareholders, its likely to earn back most of that investment, in any case.

    If investors in Tesla are the risk-takers, then why don't they demand that Tesla sell a bunch of bonds to leverage their investment? Because that's not what motivates leverage! What motivates leverage is savers looking for certainty. And, given the choice between loaning cash to Ford or Tesla, they have a clear preference for Ford.

    Think of the madness we engage in when we see a potential approaching economic contraction, and we see rising debt levels, and we react by deciding that sentiment needs to be tamped down. And, lo and behold, if we do it boldly enough, like we did in 2007 and 2008, lending actually does decline when we tear up the financial system. And we pat ourselves on the back. "See. All that risk-seeking debt led to an inevitable collapse, and now those borrowers are finally deleveraging in the way smart people like us knew they needed to." And, library shelves fill up with articles about the mystery of why interest rates remain so low after the crisis.

  6. Agree in general with your claim that we should only take large-scale economic actions based on mechanisms with known causal channels. Problem is that at this point we shade into psychology of crowds, and one fact we know there is that applying math to human behavior is notoriously tricky. Campbell's Law (AKA Goodhart's Law) comes into play and wrecks the algorithms.
    However, one quibble: borrowing at 1% and lending at 3% is not the same as borrowing at 5% and lending at 7%, and never can be, because of the nature of the investments available in a near-zero-lower-bound economy. We typically hit the ZLB only because safe investments have dried up.
    This is also why buying at 20% and lending at 23% is not even remotely akin to buying at 5% and lending at 7%. Investments aplenty may abound in a 20% inflation environment, but in such an economy things have gone so badly off the rails that it's likely anyone who gets a loan at a 23% interest rate will have trouble paying it back if there's a blip in the economy. (Fed jacks prime rate way up to kill inflation, or hyperinflation breaks out, or a major economic crash hits.)
    The point is that a 1% prime rate environment is as historically abnormal as a 20% prime rate environment. Both are indicators that something has gone badly wrong with the underlying economy. This has a huge effect on the quality of investments available.
    It may not have the support of mathematically provable equations, but Walter Bagehot's description in Lombard Street (1873) still rings true:
    "People won't take 2 per cent; they won't bear a loss of income. Instead of that dreadful event, they invest their careful savings in something impossible - a canal to Kamchatka, a railway to Watchet, a plan for animating the Dead Sea, a corporation for shipping skates to the Torrid Zone."

  7. Interesting post: working in financial markets though, i must tell you that the real issue is valuations and Debt/equity. Low rates stimulate leverage. ANd therefore the equity part becomes a smaller part of the EV. Therefore equity starts to act as a warrant on debt. It should be more volatile not less. Low vol is therefore very difficult to explain in equities, if not for the fact that companies are not gearing up exactly coz the know this level of rates is an anomaly.

  8. Valter Buffo, Recce'd, MilanJuly 6, 2017 at 4:10 AM

    I agree, and disagree at the same time, with the content of this post. I fully agree with the fact that "it is all about knowledge", and I agree about the fact that the Fed has no (or little) "technocratic competence to understand the difference between "boom" and "bubble", ". Also, I agree with the fact that the Fed should not be endorsed of any "political mandate to determine the correct level of stock prices". Fine. Then, why the Fed has been messing with the financial markets at all, for a ten-year long period, since 2008? The Fed has transformed a temporary intervention to provide liquidity to a few market participants into a 5 trillion balance sheet; further the Fed has explicitly said that her goal was and is to "reduce market risk premia": we all know that, as of today, there is very little agreement about what these risk premia are, how large or small they are, and most of all IF these risk premia exist at all (I personally believe they are just the by-product of one of the many theories for asset pricing). Those women and men at the Fed do not have the knowledge: they do not know better than you or me what the yield on a corporate bond or an MBS should be, and distorting permanently its price they simply generate a loss of welfare to the entire economic system. Final consideration: it is maybe questionable if low official rates generate higher asset prices. But there is no question that if one single market participant buys 18% of the total amount oustanding of a class of securities, prices for those securities will certainly go up, and generate a spillover effect on other asset classes, heavily distorting the price mechanism.

  9. We believe in free markets, and free markets to allocate capital.

    But you know, these hard-headed financial types go bananas when there are "low" interest rates. They become financial strumpets! Free markets cannot be trusted when interest rates are "low."

    What is "low" anyway? I suspect yields on 10-year U.S. Treasuries are set by the market, not the Fed. I got news for anybody: They may be "low" for another generation or two. See Japan.

    The guys in pinstripes will be wearing Bozo outfits soon, with all these low rates. Low rates for as far as the eye can see.

    And yet if the Fed raises short-term rates…long-term rates may go even lower.

  10. Much more likely to be circumspect with your own money. Raise the capital requirements to avoid bubblishess financial activities.

    Pete Bias

  11. Valter Buffo, Recce'd, MilanJuly 7, 2017 at 1:48 AM

    Further to my previous comment.

    The following text is from a speech of Mario Draghi, May 6, 2013

    But what are risk premia?
    The compensation required on a long-term financial contract must be at least equal to what could be obtained with a short-term contract that is continuously renewed until the end of the term. Long-term investors require a return that, at the very least, establishes financial equivalence between the two strategies. But, in general, equivalence is not enough. Creditors expect additional compensation for the risks they take on relating to not being paid back promptly. These risks are varied and the markets attach a price – or risk premium – to each one. The pure risk relating to postponing the availability of capital for a period of time is compensated by a term risk premium. The risk that a creditor is forced to liquidate a long-term financial investment before maturity in difficult market circumstances is compensated by a liquidity risk premium. Finally, the risk that the borrower does not meet his/her repayment obligations at the end of the contractually-agreed term is compensated by a credit risk premium.
    In a period of deep financial crisis, the increase in all risk premia is, as I mentioned, out of proportion because market participants are no longer willing or able to bear them.
    The non-standard measures adopted by the central banks in the larger countries in the five years since the start of the financial crisis can be identified according to the type of risk premia they were intended to tackle.

    End of quote. Question: is all this qualifiable as "a known and verified fact, ready for exploitation by policy makers"?

    I personally think not. These arguments have been exploited to justify an hazardous effort of unconventional and conservative policy, whose main goal was and still is preservation (of the financial system AS IT WAS in 2007), with heavy negative implications in terms of productivity and welfare for the real economy, and financial markets as well.

    1. That's a very interesting and revealing quote. The proposition that long term bonds are riskier than short term bonds is true from the perspective of a bank or banker, with very short horizon. For an individual with long horizon, the opposite can be true -- the safe asset is a long term (indexed) bond. For a long horizon, short term bonds carry rollover risk, while you know exactly what you get in a long term bond.

      So it is a view that only highly leveraged intermediaries ever participate in markets. Sure, they are the ones there in the instant, but never is a long time. Especially considering life insurance, pension funds, endowments and sovereign wealth funds with very long horizons, no urgent mark to market needs and no leverage.

      "market participants are no longer willing or able to bear them" is also interesting. At any price? Or at a price that Mr. Draghi, perched atop the ECB, likes? Indeed, pools of capital waiting for the next fire sale = buying opportunity are rare. But perhaps they have been burned before. Again, it is the view that only highly leveraged, high speed traders are ever present in markets. A very short - term, very institution based view, and one that ought to ask why have we scared everyone else away?

    2. Valter Buffo, Recce'd, MilanJuly 13, 2017 at 2:28 AM

      This is a very late comment that probably only you, John, will read. Nontheless, it might provide a clarification on the role of those "highly leveraged intermediaries" that you are mentioning here. As of today (2017) many comments stress the "lower level of financial leverage" in the financial industry, with respect to the 2006-2008 levels. This representation is misleading. Facts are as follows: the total level of debt in the whole economy as well as in the financial system (with respect to assets and with respect to GDP) has in fact increased after the GFS. Lower leverage ratios are obtained with referencing only to the quantity of "credit": but we all know that "credit" in traditional (banking) forms has been substituted by larger bond financing. The move towards bond financing is explained by the fact that many (even at the Central Bank!) assume that securities in general are "more liquid" than "credit" due to tradability in financial markets. A dangerous simplification. I shoud say something stronger than "dangerous", I should even say ... stupid. On bond markets, liquidity is NOT guaranteed by any "superior" Authority, it is provided solely by willing market participants: when no market participant wants to expand further its asset holdings, even at very low market prices, liquidity disappears. An it happens: phenomena of widespread market un-liquidity (domino effects) happened, many times, even in our recent financial history. Some would then oppose that the Central Bank stands ready to buy unlimited quantities of bonds: that argument, again, should be qualified as ... not very clever. A Central Bank can buy only LIMITED quantities of each category of assets, since if she buys more that (say) 35% of the outstanding amount, the market would simply cease to exist, along with its institutions. As it happened for MBS in the US. The main success of the Fed and the ECB, at least until now 2017, has been to feed the illusion that a "perfectly functioning" free market for bonds is still in existence (which is NOT). Back to "highly leveraged intermediaries", then: today, "leveraged" has a very different meaning than 10 years ago. Leverage then came from mostly bank credit: leverage today is "leverage from attributing value to potentially un-marketable securities".

  12. This is pretty odd given his most recent op-ed in WaPo where he explicitly derides the hubris of elites:


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.