March 12, 2008
Going, going, gone: a rising auction of scary scenarios
By Martin Wolf
What am I bid on financial sector losses from the US subprime mortgage crisis? Do I have advances on the $100bn suggested by Ben Bernanke, chairman of the Federal Reserve, only last July? Yes, I now have $500bn from the gentlemen from Goldman Sachs. Any advances on $500bn? Yes, I have $1,000bn-$2,000bn from Nouriel Roubini of New York University’s Stern School of Business. Any advances? Going, going, gone.
It is easy to be cynical about this ascending auction of scary prognoses. But we cannot ignore them.
In “Why Washington’s rescue cannot end the crisis story” (this page, February 27) I analysed the implications of aggregate financial sector losses of $1,000bn. That figure was in line with estimates by Prof Roubini and George Magnus of UBS.
I concluded that even this would be manageable, if painful, for an economy as big and a government as creditworthy as that of the US. Prof Roubini objects that I have taken the downside too lightly. He now argues that financial losses might amount to $3,000bn.
The remainder of this column can be read here. Debate from our panel of economists appears below.
Fernando Martel, (guest commenter)
Dear Mr Wolf,
I get the sense that both you and Prof Roubini are engaged in double counting. You either take the household’s equity losses or the bank’s write downs but not both.
Think of it as a simple production function. There is a big temporary shock to the depreciation rate of capital. We can then measure the impact of this as:
(1) Change in the value of the physical capital stock;
(2) Change in the financial value of capital (e.g. stock market value of REITS, mortgages, bank write offs, etc.);
(3) Changes in the net present value of future income.
You need to choose one, but not all three.
Second, whatever happened to housing as long-term investments? Sure, there are losses in the books and many people speculated, but not all losses will be realized if home owners do not sell. This is not to say things are OK, but just to temper some of the wildest pessimism I am seeing.
Finally, Prof Roubini predicted a recession three years ago. Is that amazing foresight or bad timing? I mean, he never gave a specific date and there is nothing more sure than recessions, death, and taxes, so if you predicted one now you are bound to be right in 0 to 7 years, more or less.
Posted by: Fernando Martel | March 12th, 2008 at 2:28 pm | Report this commentFernando Martel is a Ph.D student at NYU.
Martin, when I am confronted with the analyses and predictions for the US economy by Nouriel Roubini, George Magnus, or by David Greenlaw, Jan Hatzius, Anil K Kashyap, Hyun Song Shin in their paper “Leveraged Losses: Lessons from the Mortgage Market Meltdown” ( http://www.brandeis.edu/global/rosenberg_institute/usmpf_2008.pdf ), I am strongly reminded of the logical possibility that their predictions can be right even though their analysis is rubbish.
What bothers me most in their analysis is that the losses of the US banking sector (or (highly) leveraged sector) that they focus on and whose impact on the real economy they discuss, are to a large extent losses on inside assets, that is, financial assets for which there is an exactly matching financial liability somewhere else in the system. I quite buy the argument that if the value of a bank’s capital is put under pressure by losses on its balance sheet, the bank may well curtain lending, including lending to consumers and non-financial corporations. Such a reduction in lending to households and non-financial businesses may then lead to a reduction in consumer spending and investment respectively.
But when the losses involved are on inside assets, there is a matching winner somewhere else. Unless we analyse the implications of this gain for spending alongside the implications of the loss in the bank, we are at risk of seriously overstating the trouble we are facing.
Let me illustrate with a simple example. A more comprehensive discussion can be found in my blog as Double counting 101: inside versus outside assets ( http://blogs.ft.com/maverecon/2008/03/double-counting-101-inside-versus-outside-assets/ ).
Consider a household that purchases a home worth $400,000 with $100,000 of its own money and a mortgage of $300,000 secured against the property. A house is, of course, an outside asset – an asset for which there is no matching liability. Assume the price of the home halves as soon as the purchase is completed. With negative equity of $100,000 the home owner chooses to default. The mortgage now is worth nothing. The bank forecloses, repossesses the house and sells it for $200,000, incurring costs of $50,000 in the process.
The loss of net wealth as a result of the price collapse and the subsequent default and repossession is $250,000: the $200,000 reduction in the value of the house and the $50,000 repossession costs (lawyers, bailiffs etc). The homeowner loses $100,000, his original, pre-price collapse equity in the house - the difference between what he paid for the house and the value of the mortgage he took out. The bank loses $150,000, the sum of the $100,000 excess of the value of the mortgage over the post-collapse low price of the house and the $50,000 real foreclosure costs. The $300,000 mortgage is an inside asset - an asset to the bank and a liability to the homeowner-borrower. When it gets wiped out, the borrower gains (by no longer having to service the debt) what the lender loses.
The legal event of default and foreclosure, however, is certainly not neutral. In this case it triggers the repossession procedure that uses up $50,000 of real resources.
How does the redistribution, following the default, of $100,000 from the bank to the defaulting borrower - the write-off of the excess of the face value of the mortgage over the new low value of the house - affect aggregate demand?
There is one transmission channel that suggests it is likely, had this redistribution not taken place, that demand would have fallen more than it does following the default. The homeowner-borrower is likely to have a higher marginal propensity to spend out of current resources than the owners of the bank - residential mortgage borrowers are more likely to be liquidity-constrained than the shareholders of the mortgage lender.
Finally, we have to allow for the effect of the mortgage default on the willingness and ability of the bank to make new loans and to roll over existing loans. Clearly, the write off or write-down of the mortgage will put pressure on the bank’s capital adequacy. The bank can respond by reducing its dividends, by issuing additional equity or by curtailing lending. The greatest threat to economic activity presumably comes from new lending. This is the only channel of transmission Roubini and Greenlaw, Hatzius, Kashyap and Shin recognise.
When leveraged institutions suffer an $800bn (or $3 trillion) loss on mortgage-backed securities, there is a matching $800bn (or $3 trillion) gain on mortgage-backed securities for the issuer of these securities (say the SPV that issued them). If the SPV has the actual pool of mortgages, bought from the originators of the mortgages) on the asset side of its balance sheet, these mortgages presumably have taken an $800bn (or $3 trillion) beating. Corresponding to that is the gain to the homeowners who borrowed through these mortgages. If the mortgage borrower has defaulted and the mortgages are written off, the gain to the borrower (compared to the situation where he was still servicing the mortgage) is immediately obvious. If the mortgages are merely written down because of an increased likelihood of default by the borrower, but an actual default has not yet occurred, the borrower of course still meets the current mortgage service payments, which may put him under cash-flow pressure. It is also possible that the likelihood of default reflected in the pricing of the mortgage-backed securities is not the same as the ultimate borrowers assessment of the likelihood that he will default.
But the general point remains. When the asset whose value is tanking is an inside asset (one that is a liability of some other natural or legal person), you must analyse the implications for demand and the economy in general of the party that gains – the holder of the inside liability that is the counterpart of the inside asset.
Posted by: Willem Buiter | March 12th, 2008 at 4:29 pm | Report this commentRobert Wade: Suppose Roubini is roughly on target: for the US, a cumulative house price fall of 30%, wiping off equivalent of 40% of GDP and 10% of household wealth; plus a nearly equivalent loss of value of stocks. I want to mention a few of the salient political repercussions of losses of this magnitude.
First, it is worth repeating a point I have made before, about impacts on the working class in America. Caner and Wolff calculate that as of 1999 fully 25% of American households were in “asset poverty”: with insufficient wealth (including houses) to survive on their own by spending down their wealth in case their income flow stops (Asena Caner and Edward Wolff, “The tragedy of asset poverty in the U.S.”, Challenge, Jan-Feb 2004.) As Caner and Wolff remark, “economic and financial developments benefited only a relatively small part of the population in the United States in the years 1984-99″. The big question is how these households will react as house prices continue to fall and consumer prices continue to rise; and whether any political party will be able to mobilize them and their discontent. Or extra-political party.
Second, the middle classes also take a big hit in less-than-obvious ways. Many middle class families responded to asset appreciation over the past several decades by issuing financial liabilities against rising asset values or by selling assets, and using the revenue to finance consumption and their own private “welfare state” - private school fees, private health care, private pensions, and the like. Conversely, they supported “reforms” to cut taxes and cut the public welfare state. Now their private welfare arrangements suddenly look much less viable. Some are likely to appeal for a bigger public welfare state targeted at middle class goods and services. How will US political parties respond, especially given the intense aversion to taxation?
Third, as the crisis hits Europe — if less intensely — similar questions arise. The European context is all the more interesting because the “natural” champions of a stronger welfare state - the center-left parties - are “in retreat”, in Italy, France, Germany,
Britain, even the Nordic countries (Ernst Hillebrand, “Europe’s failing left”, Prospect, March 08). Will Europe’s center-left parties be able to seize the opportunity of the crisis to develop a new policy story, with more of a focus on distributive justice than the “third way” and with a less admiring stance towards “globalization”?
Finally, there is the question of what, going forward, should be done in policy terms about the three deep drivers of the
“financialization of the economy” (FOE) over the period since the 1970s, which have generated the financial fragility that lies behind the present crisis. They are: (1) the proliferation of lightly regulated or unregulated financial markets, which gave rise to waves of innovation in financial products; (2) funded pension schemes producing huge pools of resources seeking high returns; and (3) deregulation of housing market credit. These drivers helped to produce the tendency to asset appreciation faster than income growth. Asset appreciation in turn subverted pillars of the capitalist economy. First, it encouraged households to switch from saving to finance future consumption to boosting current consumption by drawing on capital gains - hence households’ saving went negative. Second, asset appreciation encouraged firms to issue more shares than they needed to finance their current and
future operations, and to use the excess to finance balance sheet restructurings, including mergers and acquisitions; and to focus company strategy on these activities, eclipsing the idea of investing to produce goods and services that customers want to buy - and so also undercutting a national industrial renaissance. Third, banks became more fragile as big companies stopped borrowing, and banks responded by lending to riskier customers and embracing “originate and distribute” securitizations. (Here I draw on arguments of SOAS’s Jan Toporowski.)
Governments seeking to reduce the risks of financial instability will have to rein in the big drivers; but how, when the interests defending them are so strong? For a start, central banks should be mandated to pay attention to asset prices.
Posted by: Robert Wade | March 13th, 2008 at 11:42 am | Report this commentPaul Seabright: Another reason to be cautious about the big numbers being thrown around this debate is there is a world of difference between an asset price fall that destroys real pre-existing value, and one that merely corrects for the exaggerated asset price rise that preceded it (and which therefore “destroys” value that never really existed in the first place). A fall in house prices, which still leaves the same people occupying the same houses, is not the same as an earthquake that destroys houses. There will certainly be some real value destruction arising from the current crisis - houses remaining empty because of foreclosure are as good as houses destroyed, though the situation is easier to remedy. The fall in asset prices will also not be without cost even if all houses remain occupied, since their lower price will prevent some people from switching as easily into other assets as they coudl otherwise have done. But such switches could only ever have taken place at the margin, and it makes no sense to treat the “losses” from falling house prices on a par with real economic costs of lost output. It’s worth recalling in the midst of this crisis that subprime mortgages have enabled a lot of people to live in houses that they would not otherwise have been able to occupy. This was not a benefit worth the social cost, as we are now realizing, but that doesn’t mean it wasn’t a benefit at all. I don’t know if anyone has tried seriously to estimate the real, as opposed to the paper costs of the crisis, but I would find that more illuminating than the figures currently circulating the blogosphere.
Posted by: Paul Seabright | March 14th, 2008 at 11:29 am | Report this commentMartin Wolf: I like the comments on this column for many reasons, but not least because we have a student at NYU attacking a professor at NYU and an adviser to Goldman Sachs attacking Goldman Sachs economists.
Let me respond as follows.
I agree with Mr Martel, that one cannot add the total losses on the value of housing and the losses to banks. What I meant to argue was that the latter were an important component of the former. That is to say, to the extent that losses do not fall on households, they fall on intermediaries (or, increasingly, the state). I very much agree that not all losses will be realised by the financial sector. Indeed, I tried to point out that the big difference between Prof Roubini and the Goldman Sachs authors was over how large a proportion of the losses would be realised.
Finally, Mr Martel seems to follow Prof Roubini’s work more closely than I do. I am unaware of a forecast of recession three years ago. But I am usually too early myself. So I feel sympathetic. This reminds me of the old saw: forecast a number or a date, but never both.
Willem is clearly right. He usually is. There are two big questions here. The first is whether there is a wealth effect from house prices, in either direction. The second is whether the decapitalisation of financial institutions even if their losses are, by definition, someone else’s gains, matters. Willem is absolutely right to ask people who think the current situation is dangerous to spell out the mechanisms. It is not clear to me that there are none.
In theory, there is no reason why a wealth effect from house prices need occur. There is an owner and a renter. In the case of owner-occupied housing, these are one and the same person. If house prices rise, the owner gains and the renter loses, by equal amounts, on a present value basis. But I wonder whether this is how it appears to people. They may have bet on the absolute difference in price between the house they own in the city where they work against some other cheaper retirement home elsewhere. Assume all these prices fall proportionately. Then the owner is worse off. Again, the absolute equity the owner would own would be smaller, limiting the pension he can obtain from the asset. Of course, it is also true that purchasers would find housing cheaper: they are better off. So maybe this, too, makes no difference. I regard this as an empirical question. My reading of the literature is that property prices do affect consumption. They certainly affect investment in construction.
I also have a wrinkle to offer. Let us suppose that a vast proportion of the borrowing was purely speculative. People borrowed the full value of the house. Then, if house prices went up, they would sell and take the profit. If house prices went down, they would walk away and default. These then were one-way options. The financial sector may have put these as assets on the book. But they were only assets if house prices went up. Now comes the house-price collapse. The borrowers default as they always planned to. They are no better off than before. The financial sector has to recognise the reality that they it has lost its bet. So the financial sector appears clearly worse off. The reason for this is that the financial sector either did not understand the true nature of the contract or failed to account for it properly.
More important, however, is what happens to the financial system with large-scale defaults. If households default in large numbers, the financial system as a whole will be the loser. Some institutions will be decapitalised. This surely affects their willingness and ability to lend. If that starts to affect the economy, there can be a vicious circle, with lower activity, higher unemployment, more default and so forth. I presume Willem agrees on this. At the extreme, we would have what Irving Fisher called debt deflation. I presume this is what Ben Bernanke is trying to prevent. Does Willem believe this is no danger?
The big point then is that big redistributions of wealth matter if they destroy a set of core institutions in the economy. That, presumably, is why we regulate banks.
I think Robert Wade raises an important question: what is going to happen to financial globalisation? I think we honestly do not know any more. My own guess is that the high tide of laissez faire is now passed. Governments, having been dragged in to rescue the financial system, are not going to listen to these interests telling them they do not know what they are doing. And, if governments do not behave in that way, different governments will be elected. An implicit social contract has been violated between the super-rich financiers and the rest: we will leave the former alone if they do not create too big a set of problems for the rest of us. That contract has been broken. The ramifications will be enormous.
I think Robert raises good questions about the future of “financialisation”. I will come back to that in future columns.
Paul Seabright is clearly right: we are recognising reality more than distorting it. The distortion was in the previous situation. The truth sets us free. But it can be painful to recognise it. I suppose the core real cost is excessive investment in residential construction. In the end, the houses will all be occupied. So that cost should not be too terrible.
Posted by: Martin Wolf | March 18th, 2008 at 11:36 am | Report this comment