Friday, April 2, 2010

Is Cash Expenditure Targeting a "Panacea?"

An anonymous commentator asked if I thought cash expenditure targeting is a panacea in response to financial panics. I thought I said "no," but the commentator read what I wrote as "yes."

Cash expenditure targeting does not prevent the stockholders of banks that made bad loans from taking losses. If the losses are severe enough, it doesn't protect bank creditors, including uninsured depositors, from loss either. While insured depositors are protected, the deposit insurer may also take losses. And that means that taxpayers are subject to loss one way or another. (I oppose the existence of deposit insurance, though I do not favor getting rid of it "cold turkey" ever, and certainly not when many, if not all, banks are in danger of insolvency. )

If the banking system is full of bad loans because banks have lent heavily into a speculative bubble, and that bubble had implications for the allocation of resources, then cash expenditure targeting doesn't avoid the need to reallocate resources. For example, if single family houses were overbuilt, then fewer single family houses should be produced, and instead, other goods and services should be produced. The result will almost certainly be structural unemployment for an extended period of time. Carpenters are laid off and must become waiters--or something. Similarly, capital goods specific to housing, like saw mills, will lose value. Equity investors in housing related industries must certainly take losses. If the losses are severe enough, firms may fail, and those holding debt may take losses too.

The shift in the composition of demand, away from overbuilt housing to whatever it is people really want, has reduced the productive capacity of the economy. The relevant productive capacity is a capacity to produce what people actually want to buy. Real output and employment will grow more slowly, and if the shift was severe enough, it may fall. With cash expenditure targeting, the result of this will be higher inflation for a time, an increase in the price level, slower growth in real incomes, and perhaps even lower real incomes. Nominal incomes, on the other hand, will continue to grow.

While there are surpluses of those particular goods that were overproduced, there are shortages of other things. In those areas of the economy, there are higher prices, higher profits, and both a signal and incentive to expand hiring and purchase appropriate capital goods. As workers are hired and appropriate capital goods are produced, real output will grow more rapidly. The unemployment rate will fall back to its previous level (more or less.) With cash expenditure targeting, this results in deflation. The price level falls back towards its pre-crisis level. Real incomes grow more rapidly, though nominal incomes continue to grow at an unchanged rate.

In the end, real output will return to growing as it did, the unemployment rate will fall back to where it was. The inflation rate will be zero (at least with a 3% target for cash expenditure growth.) However, it is likely that real output will be on a somewhat lower growth path and the price level slightly higher than it would have been if resources had never been misallocated by the speculative bubble.

A second type of problem can result from disruption in credit markets due to bank insolvency. Even if cash expenditures continues to grow on target, it is possible that firms that have buyers and can make profitable sales will be unable to obtain credit to finance production or delivery. This is a "supply-shock." A lack of credit forms a bottleneck for production. The result is reduced real output and employment. With cash expenditure targeting, this results in shortages of goods and services, and higher prices. The profitability of the production that can be continued, both using whatever remaining credit is provided as well as by self finance, is enhanced.

If secondary effects of disrupted credit markets are considered, the pattern of demand resulting from self-finance will likely be different than what would exists with credit markets that operate. Households consume what they earn, and firms fund production and investment with retained earnings. The movement of funds between and among households and firms through credit is disrupted. Would be lenders consume and invest more. Would be borrowers consume and invest less. Assuming that the productive capacity of the economy, both labor and capital, is appropriate to a composition of demand reflecting well-functioning credit markets, then the shift in the composition of demand reflecting more self-finance allows for less production and employment, and a higher price level.

What cash expenditure targeting does is provide that the sectors of the economy with growing demand, rising prices, and rising nominal and real profits from expanded economic activity will overbalance any shrinking sectors. It is into that environment that a new bank, or an reorganized bank, will find itself. And if there has been a disruption of credit, so that a shortage of credit is creating production bottlenecks, it is an environment where the extension of credit will be profitable.

In my view, the first best solution to a financial crisis is free banking combined with index futures targeting to keep cash expenditures on target, and then rapid bankruptcy and reorganization of insolvent banks to avoid "supply-side" disruption in credit markets. Given the status quo, the Fed should have committed to keeping cash expenditures on target, and if and when that failed, returning them to target one year in the future. And then FDIC should have reorganized insolvent commercial banks and encouraged those newly solvent banks to rapidly expand enough to replace the shrinking shadow bank sector. Allowing shareholders and uninsured debtors who funded the credit bubble to take losses, while allowing a reallocation of resources, including labor, away from housing, are necessary evils.


  1. I asked what measures you thought the government should take to counter a financial panic, such as the one that recently threatened and, to some extent, actually occurred. You claim I misinterpreted your answer, but your fuller exposition seems more of the same, suggesting (to me) that you view monetary policy--pumping up the money supply to offset the (actual and threatened) fall in velocity, keeping MV or NGDP growing at a steady 3% (or whatever) rate--as the sole appropriate policy measure.

    If the panic results from the belated recognition of poor investment decisions--with banks lending into some sort of "bubble"--then there will be a "need to reallocate resources," what Arnold Kling calls "recalculation," with an associated rise in unemployment. But you suggest no role in this process for government policy. You give no reason (and I can think of none) why the government should not simply let the private sector take care of the reallocation/recalculation.

    If many banks had made poor loans, there might be "disruption in credit markets due to bank insolvency," with firms "unable to obtain credit to finance production or delivery." As the only non-monetary governmental action you advocate, you mention expedited bankruptcy proceedings for financial institutions, to minimize the impact of such disruption. But this is really something that should already have been in place, rather than a response specifically to the crisis. (By the way, how would would-be savers respond if many banks failed? I think the most likely response would be *to patronize the sound banks*; then the disruption in the credit markets would be minimal. Perhaps, not trusting any banks, they would hoard currency; most of the bad effects of this could be avoided by expansive monetary policy. You seem to think they would give up the attempt to save, and switch to consumption; I doubt that would happen to any great extent, unless they distrusted *currency and all commodities* as potential stores of value.) So far as governmental response is concerned, you note that, even if many banks are known to be insolvent, "if there has been a disruption of credit, so that a shortage of credit is creating production bottlenecks, it is an environment where the extension of credit will be profitable." *New banks* will be started up, on private initiative; the government need not do anything (except stay out of the way).

    [Continued . . . .]

  2. [Continuation . . . .]

    You write that the panic will have "reduced the productive capacity of the economy." But since, as you also write: "The relevant productive capacity is a capacity to produce what people actually want to buy," it seems likely that the pre-panic estimate of productive capacity was an illusory over-estimate: that consumers' demand had been misestimated, that they never did demand the product mix that investors/producers were geared up to produce, the value of which was accordingly overestimated. However that may be, you suggest no particular government action to deal with the supposed decline in productive capacity.

    But I have come to doubt the meaningfulness of my original question. It presupposes a fixed, given setting, so far as the financial and political institutions are concerned; but it supplies no details about what is supposed to be fixed/given. Your posts gave me the following vague impression: Woolsey is against most of the measures taken by or proposed for the government--bailouts, TARP, guaranteeing the warrantees of domestic auto manufacturers, increasing the level of deposit insurance, "fiscal stimulus," housing tax credits, Cash for Clunkers, tinkering with the tax code, etc.; and even expedited bankruptcy is something that should have been in place all along, rather than a special measure for dealing with financial crises. He would have the government rely exclusively on monetary policy. But I now realize that you reject *all* special measures, and even such a "general measure" as monetary policy. Your proposed "index futures targeting" is not really monetary policy: it is a particular *definition* of the dollar, after which no policy actions by the government will be necessary.

    So I hereby withdraw original question.