Wednesday, September 21, 2011
A Simple but Possibly Correct Theory of the Capitalist Financial Process
I rarely bother to read grand theories of economics (or politics or history) that fit into a blog post or even a short essay. How likely is it that someone has figured out a huge idea that generations of smart, knowledgeable people have all missed?
So you can stop reading here. In my defense, I recognize that most of the elements of what I’m about to say have been worked over rather thoroughly, but I’m not aware that anyone has put them together in quite this way.
It begins with the observation that credit performs two irreplaceable functions in a modern economy, income smoothing and reallocating resources to more productive uses. (a) Incomes are uncertain, but we can act more rationally if we can plan across their fluctuations. This is true of consumers who wish to avoid a feast-and-famine lifestyle (and can’t self-insure sufficiently through their own savings) and firms whose cash flow in any given period is unpredictable. Credit enables us to take a longer view. (b) Some means is needed to move income from where it has been earned in the past to where it can best be earned in the future. If you have better investment prospects than I do and can pay me a rate of interest that exceeds what I can get out of using my money myself, we are both better off, and the social efficiency of investment is improved.
Obvious, but the reason for beginning with this is to make it clear that, from a purely economic point of view, the credit relationship does not reflect or create an imbalance of virtue. It’s not about grasshoppers and ants or the provident and the prodigal, but simply an economic transaction that, if undertaken properly, is mutually, and socially, beneficial.
Of course, the expectations of lenders and borrowers can be in error. Some borrowers are dishonest, and so are some lenders. Credit terms may have depended on forecasts that prove to be wrong. How can the financial process be protected from the risk of default? On the borrower’s side, other than various types of specialized insurance, the only reliable protection is to limit the extent of leverage, but this is difficult to assess ex ante, since an unwillingness to leverage can undermine the original purposes that credit is intended to serve. Fortunately, while insolvency may be a hardship for individual borrowers, as long as the number of insolvencies is not too great at any time the financial process as a whole can withstand it.
The lender’s side is a bit different, because a capitalist economy will develop specialized lenders—financial institutions. They attempt to shield themselves from default risk by charging a risk premium. Pooled over a large number of transactions, the payment of this premium offsets the costs of default, at least if it has been correctly priced.
The system works well, in other words, as long as the frequency of default is relatively constant and predictable. This requires that default risk be uncorrelated within risk classes—that is, at the level of individual credit contracts. That is true most of the time. There are episodes, however, when risk at this level is correlated and defaults “bunch up”. An example is a financial bubble: everyone who has borrowed based on (erroneous) bubble expectations is going to be hard-pressed to service their debt. Another example is an unforeseen business cycle downturn.
When agents in the credit market face strongly correlated risks of default, the stability of the system is threatened from both sides. (a) Lots of borrowers are threatened with insolvency. They cut back on purchases they would otherwise make, and this causes a business cycle downturn, a “balance sheet recession”. The slump itself can then exacerbate the original correlated-risk condition. (b) Financial institutions can no longer cover their cumulative default losses with risk premium receipts. They too are threatened with insolvency. Their collective responses, such as calling in loans and dumping assets to shore up liquidity (Fisher’s debt deflation), also exacerbate the original problem.
Note: financial systems are more susceptible to these threats, ceteris paribus, if they have become highly leveraged; I take this to be Minsky’s core insight. Yet the problem I’m describing is logically separate from the degree of leverage. Rather, I am interested in what might be regarded as the “real” basis for financial instability, episodes in which the mistaken expectations that cause lenders to offer and borrowers to assume unserviceable debt (ex post) are highly correlated, leading to an otherwise unlikely surge of defaults.
The first-best solution, once such an episode has occurred, is to write off the bad debt as quickly as possible. This minimizes the depressive effect that troubled borrowers have on effective demand, and it enables financial institutions to resume their normally healthy pursuit of illiquidity through productive lending. One would expect border skirmishes between groups representing creditors and debtors, each asking the other to assume more of the costs, but it is in society’s collective interest to have the parties agree to such a writedown quickly, since the gains from restarting the economic machine far exceed the costs from a suboptimal choice of writedown criteria. This expeditious resolution of debt was Keynes’ preferred approach from Versailles onward.
But there are obstacles. One is ideological: in every society there is an opprobrium attached to being a debtor; the obligation to pay is not simply one of economic contract (whose expected risk of nonperformance is priced via the risk premium) but also of moral obligation. Thus resolutions that fail to impose the maximum possible pressure on debtors will be seen as destructive of the moral order. The other obstacle is that creditors are typically much richer and more powerful than debtors. They will fight proposals to reduce the liabilities of debtors with all the resources at their disposal. Instead, they favor bailouts, which are ostensibly aimed at helping the debtors, but in fact simply sustain debt service at public expense. They limit this support, however, because of their fear of moral hazard. Bailouts of sufficient magnitude have the positive effect of keeping the financial system afloat, but they are constrained by the preferences of creditors as well as the political and fiscal space available to the state, and they do little to remove the stultifying effects of bad debt on the economic engine.
This brings us to our current condition, today.