Friday, October 31, 2008

What's going on in Inflation-protected treasuries?

Yesterday the face value yield on the 5yr inflation-protected treasury (TIP) (2.88%) was less than that of the regular 5-year treasury (2.81%). Theoretically, this should be impossible. The face value yield for the TIP is a floor value; it can't pay less even in deflation. If there's any meaningful chance for inflation then the TIP's face value should be below the 5-year. This has been part of a general drop in TIP yields to predictions of almost-zero inflation; but the drop has been most extreme in the 5-year.

Traditionally the difference between with TIP and the regular bond yield is used as an inflation estimate. The Cleveland Fed, which has published this in the past, has been forced to give up on that for the time being as the result is nonsensical.

If anybody can explain this I'm all ears. The fundamental I can think of is that the TIP has such a low face yield (0.65%) that it's almost a zero-coupon. If the government defaults late in the period that affects the TIP more than the conventional treasury. But, I've seen no indicators of an appreciable risk of default late in the period. BondGuy on CR proposed that it might be unwinding hedges selling their TIPs but to me it seems somebody should step in and buy those undervalued bonds, or at least swap 5-year treasuries for them.

Thursday, October 23, 2008

Current Strange Swaps Spreads: What do they mean?

A swap spread is the difference between the fixed rate equivalent (on the market) for LIBOR over a given period and the treasury bill rate for the same period. Normally they are very close, with the LIBOR swap slightly higher than the T-bill rate, presumably because there's a chance of default on the swap.

Recently, we have seen a very strange (historically) result. Short-term spread have been astronomical - over 4% for a short period - while long-terms spreads have been very low - down to 0% briefly yesterday. So what could this mean?

The fundamentals of the swap spread are easiest to understand in terms of an investor's choice. An investor with money to invest over a certain period could buy either a swap, from a financial institution, or a Treasury. If he buys the swap, he takes on a risk of default by the swapper; if he buys the Treasury he takes on a risk of default by the US government. An investor will also have to pay a fee to the swapper for the interest rate risk, which will normally increase with the length of the swap. For the investor to be indifferent, the following equation must hold (ignoring uncertainty):

Swap rate - interest fee - swapper default risk = Treasury rate - treasury default risk.

So the swap spread at equilibrium is

interest fee + swapper default risk - treasury default risk.

High short-term swap spreads are easy to explain: high default risk by the swappers (banks and other financial institutions). Many financial institutions have large unrecognized but likely losses and default is a real risk. But why is the long-term swap so low? Simple: although the short-term risk of a treasury default is low; long term it's fairly high. High enough, actually to outweigh both the substantial risk of swapper default *and* the interest rate risk for a 30-year swap.

Is this crazy? No, not at all. Suppose the US government was in Argentina's financial situation. Long-term LIBOR swaps might not be very different, but government bonds would be paying over 25% and the swap spread would be huge - and negative.

So current swap spreads are making a prediction: that the US government will assume so much debt that over 30 years its default risk substantially exceeds that of a typical major financial institution. While not certain, this is a rational prediction based on recent government actions. Governments worldwide have declared the intent to cover financial institution losses or give them capital, with the US having the most generous giveaway, a purchase of 125 billion in 5% preferred stock (less than half market rate) in 9 major banks.

Given the magnitude of current losses - estimates are as high as 5 trillion - if the government continues with a giveaway strategy it could certainly assume staggering losses and debt. Add in expenses to deal with the depression the credit markets are predicting and the US government could easily raise its default risk above that of current financial institutions.

I think this risk shows the government must be more cautious and efficient in its bailouts. It can't just hand out money to any financial institution that needs it, without any effort to clear the books or make investors take appropriate losses. A government struggling with default imposes enormous costs on its society. Taxes necessarily become extortionate and essential services skeletonized.

The events of the past few years have shown that we need a capable and responsible government which responds to the needs of the general population. But the markets are warning us that if we allow the government to continue unrestricted giveaways to the rich and well-connected we will lose that.

Monday, October 13, 2008

Europe Starts Facing Reality

Several European governments today pledged 1.3 trillion Euros to fix the banking problems affecting their banks. In combination with the recent blanket banking protection from Britain (not yet explicitly valued but quite large) they have now offered to cover more than 2 trillion US dollars in losses. This action means the European governments have, roughly, promised enough to fix the problem on their end. The US and Japan should be able to manage interventions on similar scales to match the worldwide problem. We now have several remaining problems to deal with:

1) What about the non-EU countries, and the countries that everyone knows can't cover their commitments? UBS and the Irish banks are still exposed, as well as many smaller banks in East Europe. We still have major busts to come unless the EU acts as a whole (and why would they bail the Swiss?)

2) Can the world provide the borrowings the governments now need? In a sense they can, but I stand by my prediction they will have to borrow long for this and so long bond rates will soar. In principle, I think this is a better way to get to the "rebuilding" state than a classic panic; but the losses and the need to rebuild remain real. To put things in context, the Swedish action, which involved the injection of only 4% of GDP came with a 6% drop in GDP. Just today's intervention is 250% as much as the Swedish. Much is in the form of deposit guarantees, which obviously won't all have to be paid out, but a recession far worse than the mid-70's recession is probably unavoidable.

3) What about the shadow banking system? They aren't covered. IMO it's good ethically for the hedge funds to go down while the classic banks are protected but we'll still see a lot of damage. One good thing about that is that hedgie collapse may provide enough financial losses for the governmental interventions to fix the rest without strangling the economy. Still though, those losses will hurt.

4) Will they clean up the banks and stop the bad lending? That's necessary to get a Swedish result (bad recession followed by recovery) rather than a Japanese result (recession without end). Initial indications from Fannie/Freddie and the UK is that the governments will continue to force lending on houses, which is a very bad sign.

All in all, this permits a controlled clean-up and will probably protect the world from uncontrolled side effect disasters like rumored halts to world shipping. However, we know will find out how competent and prompt the clean-up is. And the real losses must be faced, no matter what we do.

Thursday, October 9, 2008

Why the Central Banks Can't Fix This Credit Crunch

As mentioned in my last post, the credit market is like any other market. True prices are given by the intersection of supply and demand; the government can manipulate the price by augmenting supply or interfering with demand; but attempting to set the price beyond its ability to manipulate freezes the market (the Soviet breadline situation).

Central banks create credit by issuing money. This money goes into the economy and creates credit and debt through fractional reserve banking. The way most central banks do this is by buying government bonds, which creates a "reserve" of those government bonds they can lend out in a crisis. In the case of the US Federal Reserve, that was about 850 billion dollars.

So the reserve is 850 bb, how much is the shortfall? Well, at US bank leverage (12:1) the 850 bb in cash would become about 10 trillion in credit. At European leverage (1:30) it would become about 25 trillion. In shadow banking, we have no idea what the leverage is but it's almost certainly higher than the European. Let's estimate with the median and say there's 25 trillion in credit.

Suppose 10% of that 25 trillion is lost to the massive housing bubbles and assorted LBO, consumer credit, and retail bubbles. Remember we have to count more than the US - Eurodollar losses will count too. The 2.5 trillion loss is far greater than 900 billion. It's too much - the Fed is not going to be able to hold down real rates. Perhaps I've been too pessimistic, but even a 4% loss in the credit markets - wildly optimistic at this point - is too much for them.

Note the losses are so large the Fed may not even be able to hyperinflate out of this. If the Fed confiscated all the real value of currency by hyperinflation and issued another 850 bb for a replacement currency, it still might not have enough - a total of 1.7 trillion is still less than 2.5 trillion. And of course, a hyperinflation will generate additional real losses that would need to be covered.

The Paulson plan technique of depositing Treasuries isn't going to help. At this point the government must borrow what it lends, so no net credit. Inflation, the only out, just isn't enough. A Paulson-style plan can reallocate credit from one market to another, and perhaps that's a good idea; but overall that just raises rates in one market to lower them in another.

Why haven't we seen this before? Well, first of all, this is the worst worldwide credit loss we've seen since the Great Depression, and perhaps ever in percentage terms. Second, leverage has generally increased over the past 20 years due to deregulation, so the central bank credit reserve is a smaller fraction than before. If, for example, we were at traditional US ratios, the central bank reserves would be 8% of credit, and with a 25% total inflation the central bank could cover a 10% shortfall. If losses were only 6% it wouldn't have to inflate at all. But when the problem gets bigger and the central bank (relatively) smaller they can't handle it.

For 75 we have stood on a solid ground of (apparently) unlimited central bank powers. If we came to a dip, we could step down. But that ground is not as indestructible as we thought and the raging torrent has swept it away and put us in the water. To step down now, to try to hold credit rates below any possible market-clearing rates, is to drown. It is time to swim.

Monday, October 6, 2008

Raise the Fed Funds Rate: The Old World Order has Returned.

Right now sellers of credit won't sell to purchasers at the current price (interest rate). Were this any other market, economists would immediately say that, given the current supply and demand, the price is too low and prices (interest rates) should be allowed to go up. That was the recommendation for Soviet Russia grocery stores with their infamous lines. Increased prices reduce demand, increase supply and bring the market back into balance.

For over 70 years, however, the action of governments to credit crunches has been to increase the supply of credit by liquidity interventions. This was due to the fear of exacerbating the economic slowdown that inevitably accompanies a credit crunch. It has been so universal, and practiced so long (there is scarcely an economist alive who can remember any other way), that "increase liquidity" is a reflex response of all economists to a credit crunch. Economists have essentially forgotten that when supply is short and demand is set there are two possible actions; raise the price or increase the supply, and that normally economists recommend raising the price over increasing the supply directly.

Central banks, and the governments that back them, are mighty entities with great capacities. But like all human institutions, they have their limits. We have all become so used to the idea that governments can fill the credit supply shortfall created by low interest rates that we have forgotten they have limits and cannot supply infinite credit.

The current credit crunch has been caused by a truly massive worldwide destruction of financial capital. I have often read claims that the US housing bust, identified as the cause, isn't big enough to justify this. But there is much more than just a bust in US housing. There have also been massive real estate bubbles in many European economies, including England, Ireland, Spain, Poland, the Baltic states, and an old bubble in the Netherlands kept propped up for years. There is a major property bubble in China. There is a LBO bubble in most of the Western world, as well as extensive retail and hoteling bubbles. There is a large auto bubble in the US and large credit bubbles in the US and many other countries. Many have not yet gone bust, but the wise know at least most of them will, with vast additional losses.

I believe the losses simply exceed the ability of the current government to generate and reallocate capital. If so, no possible liquidity intervention will be able to hold credit prices to their current very low level (many real interest rates in the US are actually negative). Therefore, while governments should continue the current aggressive provision of liquidity, they should now raise interest rates until markets can once again clear given that additional (effectively subsidized) liquidity.

I say this is a return to the Old World Order because when currencies were meaningfully backed by gold and silver, governments were stringently limited in their ability to lower rates and create capital by the threat of specie flight. In those days, when a panic hit, interest rates had to go up. That Old World Order is now back; even with the considerable latitude provided by fiat currencies governments have hit their limits. Our credit is less than we might wish; we must choose which of our many competing targets will get it; and the most efficient way is by market allocation, which requires increased interest rates.

We can tolerate high real interest rates. Ask Paul Volcker. But we cannot tolerate failing credit markets. Our policymakers must accept their limits, encourge efficient allocation of limited capital, and restore public confidence by showing they understand this the realities of the credit markets.

(Hat tips to many people I've been discussing credit with, especially to evilhenrypaulson and AngrySaver. Anybody else who thinks they deserve a hat tip or link - tag me.)

Saturday, October 4, 2008

What we need for this crisis

We need to examine banks in detail, with fair valuation procedures. After this bank should be immediately recapitalized with enough government capital to keep going, with the government taking an appropriate equity share. The whole procedure should be accomplished as fast as possible. Market mechanisms will not work; revaluation without recapitalization brings disaster; recapitalization without revaluation brings trillions in losses for the taxpayer. Here's why:

In general the markets are very good at allocating capital. But sometimes the markets can go very awry, and it's possible for a central planner to surpass them. As I've expressed in my Naked Capitalism upgraded comment, I think *Paulson* will be substantially worse than even a deranged market, but that might not be true for a wiser Treasury secretary.

Markets inherently cannot correct for a major malinvestment of financial capital. Financial capital represents in the real world actions and resources allocated for coordination - obviously a very important thing. If those resources or lost via malinvestment they must be replaced, which requires high relative interest rates for financial investments. However, the losses mean that current financial investments are overvalued, which requires low relative interest rates to correct. So the market can't correct the problem.

The solution, then, must be imposed by non-market cooperative activities, which on this scale means government. The correct action is to force immediate write-downs of the bad assets and then to force recapitalization by having the government provide the capital, implicitly backed by its taxing authority and ultimately its legitimacy and guns.

The write-down could be accomplished by a Defazio-type plan (bank examiners) or approximated by a devaluation. Recapitalization could be done by a bazillion plans although I think in the context of American politics preferred stock purchase (as suggested by Stiglitz, Krugman, and many others) is the way to go. Note that you have to take *both* actions - one alone leaves part of the problem unfixed.

To be fair, if you do one, then the market *can* do the other. So if you do a DeFazio plan, real interest rates can soar to draw in capital. If you do only Stiglitz/Krugman, they can collapse to erase the post-recapitalization value of financial investments. From the viewpoint of the taxpayer, Stiglitz/Krugman without DeFazio is thus a bum deal, although it will repair the economy.

I think the reason there's near-unanimity in the blogosphere on the idea of "write-down and recapitalize", across a very wide political spectrum is simply that, on basic principles, it's the right action.

Friday, October 3, 2008

Here comes the catastrophe

The House is clearly ready to pass Paulson's horror of a bill. This bill will give effectively unlimited discretionary power over the Federal government's economic policy to a man who's been wrong on almost everything for two years in office, who got rich passing off toxic garbage at Goldman Sachs, who just asked for authority to steal $700 billion dollars, and whose plan will cause a Greater Depression by sucking all the money out of the working capital markets and using it to prop up the prices of toxic mortgage derivatives. Oh, and both Presidential candidates have gone all-out for the plan.

Could it really get any worse?

Bank of the Fed Starts Operations

Amidst all the bad news, there is some good news. The Bank of the Fed is open for business and it's helping a lot. The Fed tripled its working capital duration window on Monday and commercial paper volume in those durations (>20 days) is now running only moderately below the year's averages and about twice the rate of the last two weeks in September. The market has shifted hard to asset-backed paper, which I assume reflects a Fed collateral requirement. In the longer durations asset-backed paper is at volumes far above typical averages while all other CP markets are virtually shut down.

The A2/P2 spread is down, to "only" 2.93. Still twice that of the 74-75 recession, but probably an overestimate since the market is obviously restricted in its ability to handle that kind of paper. So perhaps the market is only predicting the worst recession since 1937 now rather than a depression.