There must be times when the Federal Reserve chairman, Ben Bernanke, feels like the Wizard of Oz — someone who’s supposed to be all powerful but who’s actually just an ordinary guy. Like now.

The American economy has arrived at one of those moments when the Fed is expected to perform miracles. Signs of a possible recession abound, despite 4.5% unemployment. Housing foreclosures are rising. Inventories of unsold new homes stand at eight months of sales, up from six last year. Manufacturing orders are weak; business investment dropped at an annual rate of 3% in the fourth quarter of 2006. Cut interest rates, the Fed is urged.

Pressure comes from Congress, Wall Street, and economists. Writing in the Financial Times last week, Lawrence Summers — Treasury secretary in the Clinton administration and an eminent economist — advised easier credit. The problem is “to avoid a vicious cycle of foreclosures, declining property values, reduced consumption demand, rising unemployment, more delinquencies and more foreclosures,” he wrote.

In popular lore, the Fed is omnipotent. With deft shifts in interest rates, it can prevent both recessions and high inflation. That notion took hold in the 1990s, when the economy enjoyed a record 10-year expansion from 1991 to 2001. Indeed, there have been only two brief recessions since 1982; by contrast, there were four from 1969 to 1982.

Well, it’s not so simple, in part because the sources of the Fed’s power are increasingly mysterious.

Technically, we know what the Fed does. It alters the “federal funds” rate — the interest rate on overnight loans between banks. It does this by buying or selling U.S. Treasury securities. By buying, it provides banks with more money; the fed funds rate drops. Selling does the opposite. But why do shifts in this tiny rate move a $13 trillion economy?

Once, answers seemed obvious. In the 1970s, the banking sector accounted for nearly half of all lending in the U.S. economy. The Fed, it was said, was increasing — or decreasing — the total amount of money banks could lend. Naturally, rates shifted on other business and consumer loans. By another theory, higher interest rates on savings accounts caused consumers to shift funds from checking accounts, where they could be spent. Consumer spending would slow or, if interest rates fell, speed up.

These traditional mechanisms are no longer so powerful.

Electronic banking has largely erased the difference between checking and savings accounts. The interest rates that matter most to the economy — on mortgages, auto loans, and business borrowing — are increasingly set in the market. Investors decide what they’ll accept on bonds and “securitized” mortgages and other loans. The banking sector represents only 23% of lending. The impact of the fed funds rate has weakened. Rates on conventional 30-year mortgages, 6.2%, are now what they were in mid-2004, despite a huge jump in the fed funds rate.

None of this renders the Fed powerless. It can still alter the economy’s available credit. But the channels of its influence are more murky, indirect, and unpredictable. It cannot steer the economy single-handedly, and many other forces — technology, business and consumer confidence, global money flows — matter as much or more. There is, however, one area where the Fed’s power is unquestioned: inflation.

The greater economic stability of the past 25 years stems fundamentally from the fall of inflation — 13% in 1980. The Fed engineered that decline, beginning with the deep 1981-1982 recession; peak monthly unemployment: 10.8%. Since then the Fed has refused to supply the extra money and credit that would feed ever-worsening inflation.

The result: calmer business cycles. Short expansions that had ended in self-defeating wage-price spirals have disappeared. Expectations that inflation will remain low have become embedded in investor, worker, manager, and consumer psychology. Long-term interest rates have dropped, mortgage rates in 1982: 15%, mainly because investors don’t need to be compensated for the rapid erosion of their money.

For the last three months, the Consumer Price Index has increased at an annual rate of 4%. So-called “core inflation,” all prices minus food and energy, is up 2.7% in the past year. “Core inflation” now exceeds the Fed’s presumed target of 1% to 2%. Increases in the fed funds rate to 5.25% now from 1% in mid-2004 aim to slow the economy just enough to cut inflation by making it harder to raise prices and wages.

Abandoning that goal prematurely, just because the economy might slip into recession, risks creating bigger future problems. To be sure, Fed actions operate with ambiguous lags. But once inflationary expectations rise, long-term interest rates would probably follow. Even a mild wage-price spiral would threaten more — not less — instability. Low and stable American inflation has underpinned America’s economic success and a stable global financial system. We ought not jeopardize them.