The writer is a professor at and past president of Harvard University. He was treasury secretary from 1999 to 2001 and economic adviser to President Obama from 2009 through 2010.

It has been joked that the letters IMF stand for “it’s mostly fiscal” rather than International Monetary Fund. The IMF has long been a stalwart advocate of austerity as the route out of financial crisis, and it annually chastises dozens of countries for their fiscal indiscipline. Fiscal consolidation — a euphemism for cuts to government spending — is a staple of IMF rescue programs. As recently as last year, the IMF’s Fiscal Affairs Department was suggesting that the United States had a fiscal gap of as much as 10 percent of gross domestic product that would require massive increases in taxes and cuts in spending.

All of this makes the IMF’s recently published World Economic Outlook a remarkable and important document. In its flagship publication, the IMF advocates substantially increased public infrastructure investment, and not just in the United States but in much of the world. It further asserts that under circumstances of high unemployment, like those prevailing in much of the industrialized world, the stimulative impact will be greater if this investment is paid for by borrowing, rather than by reducing spending elsewhere or raising taxes. Most notably, the IMF asserts that properly designed infrastructure investment will reduce rather than increase government debt burdens. Stated boldly: Public infrastructure investments can pay for themselves.

Why does the IMF reach these conclusions? Consider a hypothetical investment in a new highway financed entirely with debt. Assume — counterfactually and conservatively — that the process of building the highway provides no stimulative benefit. Further assume that the investment earns only a 6 percent real return, also a very conservative assumption given widely accepted estimates of the benefits of public investment. Then, annual tax collections adjusted for inflation would increase by 1.5 percent of the amount invested, since government claims about 25 cents out of every additional dollar of income. Real interest costs — that is, interest costs less inflation — are below 1 percent in the United States and much of the industrialized world over horizons of up to 30 years. Thus, the infrastructure investment actually makes it possible to reduce burdens on future generations.

In fact, this calculation understates the positive budgetary impact of well-designed infrastructure investment, as the IMF recognized. The calculation neglects the tax revenue that comes from the stimulative benefit of putting people to work constructing new infrastructure, as well as the possible long-run benefits that come from combating recession. It neglects the reality that deferring infrastructure renewal places a burden on future generations just as surely as does government borrowing. It ignores the fact that by increasing the economy’s capacity, infrastructure investment increases the ability to handle any given level of debt. Critically, it takes no account of the fact that in many cases government can catalyze a dollar of infrastructure investment at a cost of much less than a dollar by providing a tranche of equity financing, a tax subsidy or a loan guarantee.

When it takes these factors into account, the IMF finds that a dollar of investment increases output by nearly $3. The budgetary arithmetic associated with infrastructure investment is especially attractive at a time like the present, when there are enough unused resources that greater infrastructure investment need not come at the expense of other spending. If we are entering a period of secular stagnation, unemployed resources could be available in much of the industrial world for quite some time. Even when full employment returns, there is, as the IMF shows, a strong supply-side case for infrastructure investment at a time when public capital stocks — airports, highways and the like — relative to the size of the economy have fallen by one-third or more over the last generation.

While the case for increased infrastructure investment is almost universal — possibly excepting China, where it has been used a stimulus tool for some time — the appropriate strategy for doing more differs around the world.

In the United States, the imperative is long-term budgeting for infrastructure that recognizes its benefits, as well as its costs. There is a critical need for regulatory reform to enable projects to be approved and carried forward with reasonable speed. There is also a need for recognition that government can make a critical contribution to infrastructure investment by supporting private investments in such areas as telecommunications and energy.

In Europe, the imperative is to find mechanisms for carrying out self-financing infrastructure projects outside of existing budget caps. This may be possible through the expansion of the European Investment Bank or more use of capital budget concepts in implementing fiscal reviews. For example, it is widely recognized that privatization revenue should not normally be treated as an offset to current budget deficits when it involves sacrificing a future revenue stream. But this discipline should perhaps be relaxed when revenue is invested in growth-generating projects.

In emerging markets, the need is to assure that projects are chosen in a reasonable way based on economic benefit and to reduce lags in execution.

What is crucial everywhere is the recognition that in a time of economic shortfall and inadequate public investment, there is a free lunch to be had — a way that government can strengthen the economy and its own financial position. The IMF, a bastion of “tough love” austerity, has come to this important realization. Countries with the wisdom to follow its lead will benefit.