With signs of economic recovery all around, economists are scrambling to revise their growth forecasts, The Economist is reporting in its latest issue. The average prediction for GDP growth in America in 2002, in The Economist’s poll of forecasters, jumped from 0.6% in December to 1.7% in March. Many economists reckon that growth over the year to the fourth quarter of 2002 could reach 3-4%. This is prompting analysts to lift their profit numbers too. They still need to learn patience.

New figures due to be released by America’s Department of Commerce on March 28th were widely expected to show that profits, as measured in the national accounts, rose in the fourth quarter of 2001 for the first time in more than a year. That would still leave them sharply down on a year ago. The 22% slump in profits in the year to the third quarter was the biggest decline since the 1930s.

In the mildest recession since the second world war, it is remarkable that profits slumped so sharply. One reason is that, although real GDP barely fell, nominal GDP (the total money value of all spending) saw by far the weakest growth of the past six recessions. Corporate profits tend to be more closely linked to nominal GDP.

The boom in profits in the mid-1990s is now a distant memory. Yet it fuelled expectations that such growth would continue forever. This pushed up share prices, and encouraged firms to borrow and invest heavily. When profit margins shrank in the late 1990s, firms realised they had overinvested. They slashed investment and shareholders suffered big losses.

Investors now like to blame their losses on widespread Enron-style “creative accounting” that concealed companies’ true financial health. However, if they had bothered for a moment to look at the government’s national-accounts data on profits rather than blindly accepting everything that firms and analysts told them, they would perhaps have heard alarm bells long ago.

National-accounts profits are based largely on tax returns rather than firms’ reported profits. They cover the whole economy and use less questionable accounting principles. For example, unlike reported profits, they treat employees’ stock options as an expense, when they are exercised—just like wage costs. According to an analysis by Dresdner Kleinwort Wasserstein, the national-accounts measure of profits closely tracked operating profits reported by firms until 1997. Then a huge gap started to open up (see chart). This should have been read as a signal that something fishy was going on. The divergence remains: while national-accounts profits may have bottomed, operating profits could fall for another quarter.