NEW YORK – As the green shoots of economic recovery that many people spied this spring have turned brown, questions are being raised as to whether the policy of jump-starting the economy through a massive fiscal stimulus has failed. Has Keynesian economics been proven wrong now that it has been put to the test?

That question, however, would make sense only if Keynesian economics had really been tried. Indeed, what is needed now is another dose of fiscal stimulus. If that does not happen, we can look forward to an even longer period in which the economy operates below capacity, with high unemployment.

The Obama administration seems surprised and disappointed with high and rising joblessness. It should not be. All of this was predictable. The true measure of the success of the stimulus is not the actual level of unemployment, but what unemployment would have been without the stimulus. The Obama administration was always clear that it would create some three million jobs more than what would otherwise be the case. The problem is that the shock to the economy from the financial crisis was so bad that even Obama’s seemingly huge fiscal stimulus has not been enough.

But there is another problem: In the United States, only about a quarter of the almost $800 billion stimulus was designed to be spent this year, and getting it spent even on “shovel ready” projects has been slow going. Meanwhile, US states have been faced with massive revenue shortfalls, exceeding $200 billion. Most face constitutional requirements to run balanced budgets, which means that such states are now either raising taxes or cutting expenditures –a negative stimulus that offsets at least some of the Federal government’s positive stimulus. 

At the same time, almost one-third of the stimulus was devoted to tax cuts, which Keynesian economics correctly predicted would be relatively ineffective. Households, burdened with debt while their retirement savings wither and job prospects remain dim, have spent only a fraction of the tax cuts.

In the US and elsewhere, much attention was focused on fixing the banking system. This may be necessary to restore robust growth, but it is not sufficient. Banks will not lend if the economy is in the doldrums, and American households will be particularly reluctant to borrow – at least in the profligate ways they borrowed prior to the crisis. The almighty American consumer was the engine of global growth, but it will most likely continue to sputter even after the banks are repaired. In the interim, some form of government stimulus will be required.

Some worry about America’s increasing national debt. But if a new stimulus is well designed, with much of the money spent on assets, the fiscal position and future growth can actually be made stronger.

It is a mistake to look only at a country’s liabilities, and ignore its assets. Of course, that is an argument against badly designed bank bailouts, like the one in America, which has cost US taxpayer hundreds of billions of dollars, much of it never to be recovered. The national debt has increased, with no offsetting asset placed on the government’s balance sheet. But one should not confuse corporate welfare with a Keynesian stimulus. 

A few (not many) worry that this bout of government spending will result in inflation. But the more immediate problem remains deflation, given high unemployment and excess capacity. If the economy recovers more robustly than I anticipate, spending can be canceled. Better yet, if much of the next round of stimulus is devoted to automatic stabilizers – such as compensating for the shortfall in state revenues – then if the economy does recover, the spending will not occur. There is little downside risk.

Nevertheless, there is some concern that growing inflationary expectations might result in rising long-term interest rates, offsetting the benefits of the stimulus. Here, monetary authorities must be vigilant, and continue their “non-standard” interventions – managing both short-term and long-term interest rates.

All policies entail risk. Not preparing for a second stimulus now risks a weaker economy – and the money not being there when it is needed. Stimulating an economy takes time, as the Obama administration’s difficulties in spending what it has allocated show; the full effect of these efforts may take a half-year or more to be felt.

A weaker economy means more bankruptcies and home foreclosures and higher unemployment. Even putting aside the human suffering, this means, in turn, more problems for the financial system. And, as we have seen, a weaker financial system means a weaker economy, and possibly the need for more emergency money to save it from another catastrophe. If we try to save money now, we risk spending much more later.

The Obama administration erred in asking for too small a stimulus, especially after making political compromises that caused it to be less effective than it could have been.  It made another mistake in designing a bank bailout that gave too much money with too few restrictions on too favorable terms to those who caused the economic mess in the first place – a policy that has dampened taxpayers’ appetite for more spending.

But that is politics. The economics is clear: the world needs all the advanced industrial countries to commit to another big round of real stimulus spending. This should be one of the central themes of the next G-20 meeting in Pittsburgh.

Joseph E. Stiglitz, Professor of Economics at Columbia University, chairs a Commission of Experts, appointed by the President of the UN General Assembly, on reforms of the international monetary and financial system. A new global reserve currency system is discussed in his 2006 book, Making Globalization Work.