Thursday, November 18, 2010

Take Me Down to the Paradox City

Assume, for a minute, that Paul Krugman is 100% correct in his analysis of the economy: assume that we are indeed in a liquidity trap, and the only way to escape the trap is through higher inflation. What's the best way to create this higher inflation? Why, through inflation expectations, of course. And what's the best way to foster inflation expectations? Well, here are some ideas:

- complain endlessly about government deficits;
- claim that QE2 will debase the dollar;
- whine about higher food and gas prices;
- tout gold as the ultimate investment.

Conversely, what's the absolute worst way to create an inflationary mindset in the populace? Hmm, I don't know, but writing a New York Times column that constantly warns of the dangers of deflation while pooh-poohing claims of inflation current or future, real or imagined -- that's gotta be pretty high on the list.

So within the framework of Prof. Krugman's own model, it's the right-wing fringe that is doing what's best for America, and Prof. Krugman who's doing what's worst.

But let's not steal (credit) from Paul only to pay (credit to) Peter. The situation is in fact almost perfectly symmetric. Consider Peter, Peter Schiff that is. He is a well-known inflationista and, by his own admission, owns lots of gold and other commodities ("real assets") on behalf of his clients. Yet he's constantly berating the Fed for its easy dollar policy, and calling for tighter monetary and fiscal conditions. Again, it would seem that Schiff's antagonists are doing what's best for his portfolio, while he himself is doing what's worst.

What a wonderfully paradoxical world we live in!

Thursday, November 11, 2010

Positive Feedback, for a change

Nattering nabobs of negativity? Nyet! Instead of finding further fault with economists who I disagree with, let me share some interesting articles from economists who I agree with.

First, Jim Hamilton reports on a beautifully simple test for inflation: the rolling correlations between various commodity prices. It turns out that these correlations have increased quite significantly in recent months and years. This is strong suggestive evidence that there is a common factor driving all prices higher. What could that factor be? Hmm, I wonder.

Second, Steve Waldman has a great post on the role of morality in economics. Is there such a role? He says yes: “The thing is, human affairs are a morality play, and economics, if it is to be useful at all, must be an account of human affairs”. I couldn’t agree more. There’s a follow-up post here.

Third, Raghu Rajan has two policy suggestions that are exactly the same as my own: “The US should dial back its aggressive monetary policy, focusing on repairing its own economy’s structural problems, while emerging markets should respond by allowing their exchange rates to appreciate steadily, thereby facilitating the growth of domestic demand.”

Monday, November 8, 2010

Contra Krugman, Continued

A few follow-ups to my previous post.

First, of course, the obvious question: why does any of this matter? So what if Government Gus has to pay a slightly higher interest rate on his debts? Won’t the benefits of his spending (higher employment, income and consumption, which combine to help the private sector clean up its balance sheet) outweigh the costs?

I don’t think they will. And the reason, as usual, is positive feedback. If the tipping point is reached, Gus won’t have to pay a ‘slightly higher’ rate on his debts; he will have to pay a rate that is significantly higher. High enough to burst the bond bubble and send foreign investors running for cover; high enough to send the dollar plummeting; high enough to put into serious doubt the government’s ability to roll over its debt. The US could end up in a situation like Greece. Worse, in fact, given the size and importance of the US economy, the US bond market and the US dollar.

The bursting of a bond bubble is far more dangerous than the bursting of an equity bubble or a real estate bubble. Immediate consequences include massive benefit cuts (social security and medicare), large tax increases and high inflation. Delayed consequences include steep declines in the standard of living, social unrest, and possibly war. Any policy that increases, however minutely, the likelihood of such outcomes should be considered very very carefully: is it really worth it? And in the case of QE2, I don’t think it is.

Second, I have just read through several hundred NYT blog posts by Prof. Krugman, and I notice that he has (explicitly or implicitly) addressed several of the arguments I made in my previous post. I would like to counter his points, one by one.

But before I do so, I want to make it clear that there is no personal or political animus behind my current stance. For what it’s worth, I was and am a great admirer of Prof. Krugman, as an economist and as a communicator. And I suspect that my policy baseline is very similar to his (pro free markets but with a strong appreciation of their limitations). In many cases and on many topics I agree with everything he writes.

Having said all that, I think he is dead wrong when it comes to the bond market and how it interacts with optimal government policy. And since that is a subject on which I consider myself an expert, I feel duty-bound to comment, at length. Hence my current series of blog posts.

On to Prof. Krugman’s points, in italics below.

“I base my argument on fundamentals, not market signals”

A bit of background: some critics (not me) have pointed to the fact that Prof. Krugman did not believe in the accuracy of market prices in 2006 (at the peak of the housing bubble), but seems to believe in their validity today (when it comes to interest rates). Prof. Krugman’s response to this criticism is that in all cases he is building from fundamentals, and not relying on market signals.

Fair enough, is what I say. There’s no rule that states you have to agree with the market all the time, or for that matter disagree with the market all the time. Sometimes you may think prices are justified, at other times you may think they’re incorrect. That’s fine.

What’s not fine, however, is to then use market prices as ‘additional support’ for your fundamental thesis. If you’re building from fundamentals, then fundamentals should be all you talk about. It’s not fair to constantly cite (as Prof. Krugman does) the low level of interest rates as support for your position that the market is unconcerned about deficit spending. Especially if at other times you are willing to discount market prices since they don’t conform to your position. This I think is hypocritical.

“QE2 does not materially affect the path of future deficits”

I agree that QE2 as currently envisaged won’t really affect deficits; it’s too small (or, equivalently, the current level of deficits is too big!). For the same reason, I think QE2 will be largely ineffective when it comes to its primary purpose, which is boosting spending1.

But that’s irrelevant. The relevant mechanism here is positive feedback operating on traders with short horizons, who know that they’re in a Keynesian beauty contest. The actual level of deficits matters less than the psychological perception thereof.

This of course leads neatly into Prof. Krugman’s next point:

“Explanations that invoke market psychology are worthless”

With all due respect to Prof. Krugman as a brilliant academic economist, I think I’m much better qualified to judge the value of psychology in markets than he is. And speaking with the experience of over a decade as a (fairly successful) hedge fund trader, I have to say that Prof. Krugman is simply wrong. Psychology, herding, momentum, feedback, call it what you will: it matters, it really does.

Again, note that by invoking ‘psychology’ I am not invoking ‘irrationality’. It is perfectly rational for traders in a Keynesian beauty contest to care about other people’s (no doubt subjective) perceptions of value; indeed, that’s the whole point of Keynes’ argument.

“Critics have their own personal or political agenda”

I addressed this in my preamble; let me add here, for the record, that I am not a US citizen, do not live in the USA, and do not pay US taxes. My personal stake in US fiscal / monetary policy is minimal. However, I do believe that a strong and healthy US economy is in the best interests of the world at large, and I would like to see such an outcome eventuate. Political economy need not be a zero-sum game.

“Critics were wrong about the dotcom bubble and the housing bubble; why should we listen to them now?”

There is a whiff of ad hominem in this point, nonetheless I think it’s justified; after all economics is an inexact science with no impartial arbiters, and there are lots of hacks out there who are either incompetent or malicious or both. I think it’s fair to ask ones’ critics what their credentials are, and if they have a track record of being consistently and risibly wrong, then it’s fair to ignore them.

For what it’s worth, my first ever professional trade, shorting USD swap spreads in 2000, was based on the macroeconomic conviction that the dot-com bubble would burst, pushing tax receipts lower and necessitating increased Treasury issuance. In 2001 I predicted (not online, unfortunately) that low interest rates would lead to a housing boom and potentially a housing bubble. In 2004 I set my first small shorts in homebuilder stocks. I added to these over the next few years; by 2007-08 I was short Toll Brothers, Fannie Mae, Freddie Mac, Citibank and Goldman Sachs. Every one of these trades made money. I have made many mistakes in my time as a trader, but failing to recognize the dot-com and housing bubbles was not one of them.

Of course, I could still be wrong about what happens next; I could be imagining a bond bubble where none exists. Past performance is no guarantee of future returns. But at any rate I think I deserve to be taken seriously, as a credible analyst of markets.

“Inflation is not a worry; core CPI is at just 0.8% yoy”

This point is usually bracketed with some snide remarks about inflationistas always being wrong; this time, I will ignore the ad hominem, and confine myself to four points in reply:

First, this is a statement about the past and the present, not about the future, and thus has limited predictive power, especially with respect to a measure as driven by expectations as is inflation. It’s equally relevant (or irrelevant) to point out that just 2 years ago, all-items CPI was running at 5% yoy.

Second, core CPI (that is, CPI excluding food and energy) is a terribly flawed measure. Food and energy prices may be volatile, and demand for them may be price-inelastic, but if they show a decade-long secular uptrend, then they should be included in any accurate inflation index.

Third, even non-core CPI is suspect, because of all the hedonic and other adjustments that have been made over the years. Almost every adjustment seems to have biased CPI downwards. Is it mere coincidence that the entities that publish these numbers have an incentive to keep them low? I think not.

Fourth and most important, I believe the macro dynamics at play make CPI (whether core or all-items, whether hedonically adjusted or not) an irrelevance. The danger is not CPI inflation, it is asset price inflation. As I wrote on this blog some time ago:

China has exported deflation, but this has been concentrated in very specific segments of the economy: in the prices of retail goods, and in worker salaries. It so happens that these segments are precisely the ones captured in standard measures of consumer inflation. Central bankers, lulled by this quiescence in measured inflation, have time and again erred on the side of loose monetary policy, leading directly to the asset price bubbles that have done so much harm in recent years.

Or to put it another way: China supplies certain items that are mostly included in CPI (labor and goods); China demands certain other items that are mostly excluded from CPI (food, energy, and assets – mainly bonds). Naturally, this distinction affects the quoted level of CPI. Ignoring this distinction means fundamentally misunderstanding the macro dynamics at work today.

“Critics have no coherent macro model of their own”

Actually, I do have a macro model of what’s going on, and I think it’s very similar to Prof. Krugman’s own. I am happy to concede that the US is in a liquidity trap similar to that of 1990s Japan, and I concur fully with the analysis in Prof. Krugman’s classic paper addressing the Japanese experience.

Where I differ from Prof. Krugman is in the policy implications of his analysis. Specifically, I disagree strongly with this paragraph, which many would call the key conclusion of that 1998 paper:

The way to make monetary policy effective [in a liquidity trap] is for the central bank to credibly promise to be irresponsible – to make a persuasive case that it will permit inflation to occur, thereby producing the negative real interest rates the economy needs.

I have no doubts that such a strategy would ‘work’. But the thing is, such promises have a way of getting out of hand. You can’t ask for ‘slightly higher’ inflation – say, from 1% to 4% – and expect matters to stop there. Inflation is all about expectations, and once expectations become unanchored, the sky’s the limit. Arthur Burns and William Miller (successive Fed chairmen in the 1970s) both tried to foster temporary ‘slightly higher’ inflation in order to boost employment. Both failed miserably: unemployment stayed stubbornly high while inflation skyrocketed. It took Paul Volcker and two deep recessions to bring it back under control.

I think this a clear case of the cure being worse than the disease. But then, what’s the alternative? This brings us to Prof. Krugman’s last and best point:

“Critics have no policy solution of their own”

Okay, so I don’t believe that the Fed should promise to irresponsibly deliver inflation. And I don’t believe that the Treasury should spend trillions of borrowed dollars to bail out the consumer. So what do I believe? How do I propose to get the economy out of its current recessionary mire?

Sadly, I don’t think there’s an easy answer. Economic policy is not a magic bullet; it cannot achieve the impossible. And to expect that 300 million Americans can be returned to the same patterns of income, employment and consumption that obtained before the crisis, with no adverse long-term consequences, is to expect the impossible.

Why so? Because those patterns were always unsustainable. For many years, the United States has been consuming more than it produces; importing more than it exports; spending vast sums on unnecessary wars; and spending not enough on education, infrastructure, research and technology. These deep structural imbalances were allowed to persist by a complex global financial web, which was in turn driven by the short-term incentives of participants around the world (from the Wall Street mortgage machine, to Bretton Woods II and the Asian accumulation of Treasury debt). But ‘if something cannot go on forever, it will not’. In 2008 we saw the first unravelings of that global financial web. The process picked up momentum (positive feedback as usual), and before we knew it we were in the Great Recession.

This unraveling is not something that can be easily reversed, or perhaps reversed at all. And perhaps it should not be. After all, we have been here before. When the dot-com bubble burst, the Greenspan Fed eased monetary policy to an unprecedented degree, in order to avoid short-term pain. The result was an even bigger bubble, in housing. And now that the housing bubble has burst, Prof. Krugman wants the Bernanke Fed / Geithner Treasury to break new frontiers in easy policy, again to avoid short-term pain. The result will be a still bigger bubble, in bonds. You can’t play this game forever; sooner or later the piper will have to be paid. And every new bubble inflicts more pain than the last, when it eventually bursts (see the fourth paragraph of this very blog post).

So my policy solution, insofar as I have one, is to embrace the unraveling. The recession won’t last forever. Over time, I expect to see some combination of a lower dollar, higher inflation and lower real wages in the United States. These will lead to a lower standard of living and higher exports. And these two consequences are the key: it is these, and these alone, which can see the United States break out of its rut. A lower standard of living and higher exports are necessary to repair private sector balance sheets, bridge the gap between production and consumption, and plug various deficits. But it will take a long slow grind to get there; we could easily see a decade or two of secular stagnation, a la Japan.

It’s not what I would have wished for. But it’s the least bad option.


# 1In fact, you could make a strong argument that for QE to be effective at all, it has to materially affect deficits; this is akin to the idea of being ‘credibly irresponsible’ which I shall address below. (No, Virginia, Ricardian equivalence does not hold in the real world).

Tuesday, November 2, 2010

Liquidity, Solvency, Switching Equilibria -- and Paul Krugman

Paul Krugman believes that we are in a liquidity trap (interest rates can’t go any lower), and that the only way to break out of it is via increased government spending. Here’s a column expressing his views (apologies for the long excerpt but I want to make sure I don’t quote him out of context):
One of the common arguments against fiscal policy in the current situation – one that sounds sensible – is that debt is the problem, so how can debt be the solution? Households borrowed too much; now you want the government to borrow even more?

What’s wrong with that argument? It assumes, implicitly, that debt is debt – that it doesn’t matter who owes the money. Yet that can’t be right; if it were, we wouldn’t have a problem in the first place. After all ... one person’s liability is another person’s asset.

It follows that the level of debt matters only if the distribution of net worth matters, if highly indebted players face different constraints from players with low debt. And this means that all debt isn’t created equal – which is why borrowing by some actors now can help cure problems created by excess borrowing by other actors in the past.

To see my point, imagine first a world in which there are only two kinds of people: Spendthrift Sams and Judicious Janets ... In the past the Sams have borrowed from the Janets to pay for consumption. But now something has happened that has forced the Sams to stop borrowing, and indeed to pay down their debt. For the Sams to do this, of course, the Janets must be prepared to dissave, to run down their assets. What would give them an incentive to do this? The answer is a fall in interest rates. So the normal way the economy would cope with the balance sheet problems of the Sams is through a period of low rates.

But what if even a zero rate isn’t low enough; that is, low enough to induce enough dissaving on the part of the Janets to match the savings of the Sams? Then we have a problem.

What can be done? One answer is inflation ... But what if inflation can’t or won’t be delivered?

Well, suppose a third character can come in: Government Gus. Suppose that he can borrow for a while, using the borrowed money to buy useful things like rail tunnels under the Hudson. The true social cost of these things will be very low, because he’ll be putting resources that would otherwise be unemployed to work. And he’ll also make it easier for the Sams to pay down their debt; if he keeps it up long enough, he can bring them to the point where they’re no longer so severely balance-sheet constrained, and further deficit spending is no longer required to achieve full employment.

Yes, private debt will in part have been replaced by public debt – but the point is that debt will have been shifted away from severely balance-sheet-constrained players, so that the economy’s problems will have been reduced even if the overall level of debt hasn’t fallen.

Prof. Krugman is correct as far as he goes. Increased (deficit-financed) government spending can indeed boost incomes, employment and consumption, and thus help the private sector clean up its balance sheet. Then when the private sector has recovered, government can unwind its fiscal expansion and return to running a primary surplus, thus paying down the (temporary) debt it has accumulated.

So all is hunky-dory? Not quite. We have skipped over a very crucial assumption in Prof. Krugman’s argument: the assumption that Government Gus is a ‘player with low debt’, and can thus borrow money at low interest rates, both now and in the future, for as long as it takes until the private sector has recovered and primary surpluses can be reinstated.

Is this assumption valid? Well, it certainly appears valid at the moment. After all, 30-year bond yields (the rate at which the Treasury finances its debt) are close to multi-year lows. Surely if the market were truly concerned about the size of government debt it would charge a higher interest rate on government bonds?

Not quite. Appearances can be deceptive. In this particular case, there are two errors that contribute to the mistake: a confusion between liquidity and solvency, and a misunderstanding of how markets work. Let’s take a deeper look at both of these.

Consider, first, the aforementioned two measures of fiscal position: solvency and liquidity.

Solvency is a ‘stock’: if your total assets are larger than your total liabilities, then you’re solvent.

Liquidity is a ‘flow’: if your monthly income is greater than your monthly expenditure, then you’re liquid.

Put another way, solvency is the level of your “assets minus debts” while liquidity is the derivative (the rate of change) thereof.

Now, you might think that solvency is somehow more important than liquidity, when determining the future of an enterprise. But this is not necessarily the case. You can be insolvent but stay in business for a long time, as long as you have the cash with which to pay your monthly bills. Conversely, you can be perfectly solvent but go out of business, if you don’t have cash in hand when the bill-collector comes calling.

This is a key point. Bankruptcy is always and everywhere a liquidity event. Nobody ever goes bankrupt because their consolidated balance sheet shows greater liabilities than assets. No, they go bankrupt because they can’t make a payment in the here and now.

Does that mean solvency is unimportant? Again, not quite. Because solvency and liquidity are not independent; they feed into each other. Specifically, it’s usually the perception of solvency (or insolvency) that drives the availability (or absence) of liquidity.

If you appear long-term solvent, you can often borrow money to tide you over any short-term liquidity problems. What’s more, the interest you pay on the borrowed money will be low, reflecting your perceived status as a ‘safe’ borrower, and thus enhancing your solvency. High credit quality is thus a self-fulfilling prophecy. (Once again, it’s our old friend positive feedback!)

And the converse is also true. If you appear insolvent, it’s hard to borrow money except at punitive rates; the high cost of servicing your debt further degrades your solvency, and before you know it you are in a ‘liquidity death spiral’ and out of business.

How does this apply to Government Gus? Well, there’s no question that Gus has ready access to liquidity; that is the message of the capital markets (30-year bond yields at 4%). But Gus is most definitely insolvent:
The fiscal gap is the value today (the present value) of the difference between projected spending (including servicing official debt) and projected revenue in all future years ... Based on the Congressional Budget Office’s data, I calculate a fiscal gap of $202 trillion, which is more than 15 times the official debt. This gargantuan discrepancy between our “official” debt and our actual net indebtedness isn’t surprising. It reflects what economists call the labelling problem. Congress has been very careful over the years to label most of its liabilities “unofficial” to keep them off the books and far in the future.

(The quote is from BU’s Prof. Lawrence Kotlikoff, who specializes in inter-generational accounting.)

Prof. Krugman elides the distinction; he claims that because Gus is liquid right now, he will have no problem borrowing money for as long as it takes to help the economy recover; and long-term solvency simply does not matter.

But it does. And to understand why, you have to understand how markets work.

See, traders don’t really care about long-term solvency per se (or long-term anything, for that matter). Traders only care about short-term price action, because that’s what they get paid for. Annual bonuses depend on annual profits.

And short-term price action is a mess of positive and negative feedback, self-fulfilling prophecies, tipping points and overshooting.

Right now, 30-year bond yields are low because we’re in a feedback loop, wherein banks borrow money from the Fed at 0% and deposit it in Treasury bonds at 4%. The low level of interest rates convinces everyone that government insolvency and inflation are nothing to worry about. And that in turn justifies further buying of bonds.

Thus we are in what appears to be a stable equilibrium. But we could easily flip over into another less healthy equilibrium. If, for whatever reason, the market loses some of its confidence in the government’s ability to roll over its debt, then positive feedback will begin to operate in the opposite direction1. Rates will go higher, debt service payments will go higher, and the liquidity/solvency situation will get worse, causing rates to go higher still. We saw an example of precisely this dynamic in Greece earlier this year.

Another way to look at it is this. There is no ‘right’ and ‘wrong’ when it comes to trading; no ‘fundamental value’ or ‘fair price’; there is only ‘profitable’ and ‘unprofitable’. And the sole determinant of whether a particular trade is profitable or not, is the behavior of other traders. (Keynes’ famous analogy of the beauty contest). If the market as a whole believes that Government Gus is liquid, then the optimal strategy for any given trader is to lend money to Gus and take home the carry; if the market as a whole believes that Government Gus is broke, then the optimal strategy is to squeeze him. And the belief system of the market as a whole can turn on a dime.

This is what Prof. Krugman does not seem to understand: the rapidity with which markets can go from one equilibrium to another, once a tipping point is reached, thanks to feedback effects. And the worse the initial solvency position of the government, the more likely it is that the tipping point will be reached. This is the link between solvency and liquidity; it has nothing to do with fundamentals, and everything to do with trader behaviour in a feedback market.

So we’re in a race against time. Will deficit spending resuscitate the private sector before the debt service tipping point is reached? That, indeed, is the question, and it is one that Prof. Krugman does not even ask, let alone answer.

(Personally I’m pessimistic on both fronts. I doubt that government spending will suffice to repair private sector balance sheets or confidence any time soon; Japan is Exhibit Number One for how an economy can remain in the doldrums for years despite massive government spending2. And I think the tipping point may come sooner rather than later, simply because there seems to be an utter lack of political will to do anything at all about the deficit situation. I hope I am proven wrong.)


# 1You don’t need a complete loss of confidence; all you need is a change on the margin, and positive feedback takes care of the rest.

# 2I believe Prof. Krugman’s response to the ‘Japan critique’ is to say that Japan’s own implementation of QE was not aggressive enough. There are two problems with this response. First, of course, it is a classic unfalsifiable argument. Second, and more important, I believe that one should defend or attack QE as it is, not QE as it ‘should be’. Perhaps in an ideal world we would get a sufficiently large QE to in fact solve the economy’s problems just like Prof. Krugman hopes. But in the real world the amount of QE that is politically tenable is likely to fall well short of this ideal amount. And there is no reason to expect that the effectiveness of QE scales monotonically with size; it is entirely possible that ‘no QE’ will be better for the economy than ‘half-hearted QE’, albeit worse than ‘ideal QE’.

Monday, November 1, 2010

Month-End Recap, Oct 2010

This is a new feature: a very quick summary of my investment positions, to be published monthly. This is not investment advice and should not be construed as such.

I am currently 93% invested, as follows:

59% Agricultural commodities
11% Other commodities
17% Emerging market equities
6% Miscellaneous equities

My largest position is in soybeans, followed by sugar.

I usually try to be between 85% and 105% invested, so I am a touch below my average exposure here. Indeed I plan to drop down to around 90% in the next few weeks or months, mainly through selling EM stocks especially India.