Whenever Fed Chairman Alan Greenspan testifies before Congress, as he did on Wednesday, legislators are fascinated with what they can get him to say about fiscal policy -- budget deficits. Stock and bond investors only listened to what he had to say about monetary policy. Stocks and bonds did not rise that day because investors thought Greenspan's familiar anxieties about budget deficits were newsworthy or bullish.

Why do investors pay no attention to Greenspan's warnings about budget deficits? Because they know these warnings are based on archaic theoretical conventions that have recently been well tested and found false.

Greenspan described deficits, for example, as making "demands on national savings." The idea is that government borrowing must be subtracted from an otherwise fixed amount of saving. Proponents of this idea imagine that if tax collectors would simply take more money from the private sector and give it to the government, the sum of both public and private budgets would be magically improved. It is on the basis of this sort of imaginative bookkeeping that Greenspan and others once predicted that moving from deficits to surpluses would greatly increase the "national savings rate" (public and private saving as a percent of GDP).

Did moving to surpluses from 1998 to 2001 really raise the savings rate? The Fed's Policy Report to the Congress says it did: "The federal government had contributed increasingly to national savings in the late 1990s and 2000 as budget deficits gave way to accumulating surpluses." Those words sound comforting, but the facts are not: From 1998 to 2001, the budget was in surplus and national savings was 17.5 percent of GDP. From 1981 to 1989, when budget deficits averaged 3.8 percent of GDP, the national savings rate was higher -- 18.2 percent.

The Policy Report insinuates that budget deficits caused the savings rate to fall in 2001-2002. Nice try. In reality, recessions cause budget deficits and also shrink the sources of savings (profits and jobs). Savings rates are always lower in the wake of recessions -- 14.7 percent in 1993, for example, and 14.6 percent rate of 2002.

The report graphs "net" savings (after subtracting estimated depreciation) as a percent "gross" domestic product (with depreciation added back in). Yet even that dubious "net" savings rate was still smaller when the budget was in surplus (5.6 percent in 1998-2001), than when deficits were large in 1981-89 (6 percent).

Comparing net savings to gross product is a common but disreputable statistical gimmick. It would be more honest to express net savings as a percentage of net national product. But doing that would add nearly a percentage point to the net savings rate during the Reagan years. The only purpose of comparing net savings to gross product is to minimize measured savings and exaggerate its decline. What looks like a secular slide in net savings is largely a secular acceleration in estimated depreciation. Besides, even an honest estimate of net savings would be inappropriate for the most infamous prediction of conventional theory -- twin deficits.

Greenspan still claims, "The current account deficit and the federal budget deficit are related." That is, the current account is thought to depend on the gap between gross investment and saving. Since deficits were assumed to lower the savings rate, that is why budget and current account deficits were assumed to be twins. If we would get rid of the budget deficits, Greenspan and others promised in the early 1990s, then the current account deficits would vanish too. What happened?

In 1991, the budget deficit was 4.7 percent of GDP, but the current account was in surplus. Trade warriors should beware of getting what they ask for: The reason imports fell more than exports was that the U.S. recession was worse than abroad.

After that bad start in 1991, twin deficits predictions became worse and worse. Every year the budget deficit grew smaller and smaller, before turning into rising surpluses. Yet the current account deficit grew larger and larger. By 1998, the budget surplus equaled 0.8 percent of GDP but the current account deficit was 2.5 percent. By 2000, the budget surplus was 2.4 percent of GDP, but the current account deficit was 4.4 percent. Continued...