At this writing, the economy is still in a severely depressed state, with the unemployment rate at 9.4 percent (the "underemployment" rate, which includes workers who have stopped looking for jobs and workers who are involuntarily working only part time, has risen to 16.4 percent), with a continuing surge in the personal savings rate (now 5.7 percent, up from 4.2 percent a month ago and 1 percent at this time last year), and with other shoes expected to fall soon: massive credit card defaults, a sharp decline in commercial real estate values, and an accelerating number of home foreclosures. Many economists believe that, with the economy so depressed, inflation is not a serious risk.

To evaluate that belief requires distinguishing between the short run and the long run--more specifically between a depression period (a bust) and a recovery period (a boom). Policy prescriptions tend to reverse at the ends of the business cycle. In a boom, thrift is good policy, to prepare for the inevitable bust, but in a bust, thrift is bad policy, because it reduces consumption and hence production; that is why the increase in the personal savings rate at this time is ominous. In a boom, inflation is a bad policy, because (among other things) it creates, as we know, asset-price bubbles. But in a bust, inflation is a good policy. This is partly because the biggest risk in a bust is a deflationary spiral, when as a result of falling prices the purchasing power of the dollar rises. Debts now become a crushing burden, because they are fixed in nominal terms and thus increase in real terms when they have to be repaid in dollars worth more than dollars were when the money was borrowed. And hoarding rises, because in a deflation the purchasing power of money increases even when it is just sitting in a safe-deposit box. Interest rates soar in real terms: if prices are declining at a rate of 2 percent a year, a nominal interest rate of 6 percent becomes a real interest rate of 8 percent, because a 6 percent increase in the number of dollars one has is the equivalent of an 8 percent increase in purchasing power. In effect, then, idle cash earns 2 percent interest. And the more dollars that are hoarded, the less economic activity there is.

Inflation can offset deflation. In the example I just gave, a 2 percent inflation rate will eliminate the deflation. But a higher inflation rate would be even better (within limits), because inflation operates as a tax on cash balances, and hence on hoarding, and the higher the rate of inflation, the heavier the tax and so the less hoarding. Suppose as before that deflation is 2 percent, and the Federal Reserve manages to increase interest rates by 4 percent. Then cash balances will erode by 2 percent a year, and this will induce the hoarders to put their cash to work, either in consumption or in active investment.

Moreover, inflation reduces the real wages of workers by raising prices, while if prices are falling real wages will be rising even if workers receive no raises: the same dollars will buy more goods and services. A reduction in real wages reduces labor costs, which encourages companies to do more hiring, and thus reduces unemployment. Workers in a depression will tend to accept a reduction in their real wages because they fear being replaced by what Karl Marx called "the reserve army of the unemployed."

Perhaps most important in the current economic situation is that, as I have already noted, inflation reduces the burden of debt. This is important for banks, but also for homeowners who have fixed-payment mortgages (as opposed to adjustable-rate mortgages). Even if a homeowner's nominal wage does not rise--and therefore, if there is inflation, his real wage falls--if inflation causes the market value of his house to rise he will have greater equity in the house and so will be less likely to default. 

So a moderate inflation can speed recovery from a depression. The problem is that it can be difficult to create inflation in a depression. The Federal Reserve can pump cash into the economy by buying Treasury bonds and other debt, as it is doing; but if the sellers decide to hoard the cash they receive from the sales, as the banks are largely doing; the injection of cash will not raise prices. I have noted in previous blog entries that the banks now have more than $800 billion in "excess reserves," which means cash that they are permitted by the regulatory authorities to lend but that instead they are retaining as cash (or the equivalent--mainly credit accounts in federal reserve banks). Money that doesn't circulate doesn't increase the ratio of dollars to output and therefore does not increase prices.

But additions to national debt, as a result not of "easy money" but of budget deficits--spending not offset by tax revenues--can create an expectation of future inflation (the debtor's friend, even when the debtor is a government). So can a huge amount of hoarded cash, which when the economy recovers may be lent or otherwise spent in a short amount of time, thus increasing the ratio of money in circulation to output.

The prospect of future inflation will not increase prices immediately, and it will not affect short-term interest rates, but it will increase long-term interest rates. And this seems to be happening. The ten-year Treasury bond interest rate has been rising rapidly and is now closing on 4 percent. Long-term interest rates for private debt, such as mortgages, are rising to even higher levels because of the risk of default on private debt (no one think the federal government is going to default on its debt, huge as it is). The 30-year mortgage interest rate is closing on 6 percent. This bodes ill for the recovery of the housing market, because it prevents the refinancing or mortgages and reduces the sale price of housing, since housing is cheaper the lower mortgage interest rates are.

One might think that an increase in interest rates that was due to an anticipation of inflation would have no effect on economic activity, including housing prices, because borrowers would expect to be paying the higher interest rates in cheaper dollars. But this ignores liquidity constraints. If you want to refinance your mortgage, or to finance the purchase of a house with a fixed-payment mortgage, and the mortgage-interest rate has risen, you will have to shell out more cash each month, and you may not be able to afford to do that.

It is not certain that long-term interest rates are rising because of inflation. They may be rising simply because the demand for long-term debt is rising. The government is borrowing more to finance its growing deficit, and there are signs that long-term borrowing for investment projects is also increasing. But whether inflation or demand is responsible for rising long-term interest rates, the rise is a negative from the standpoint point of speeding economic recovery. This observation is not in conflict with my earlier argument that we want some inflation in an economic downturn. For we must bear in mind the difference between short run and long run. Inflation in the short run, which we do not have and which the Federal Reserve may not be able to create, would be good, but inflation in the long run is bad, both in general and, because of the effect on mortgage rates, with specific reference to our current economic situation. 

We want to hear what you think about this article. Submit a letter to the editor or write to