Here are four things you often hear asserted about the economy. Guess which ones are wrong:

l The major reason job growth has been so slow is that there wasn’t enough fiscal stimulus.

l  Off-shoring creates more jobs at home than it kills.

l Industrial policy always fails.

l Globalization is not a major contributor to rising income inequality.

In fact, all four statements are wrong. And now Michael Spence has the proof.

Spence is former dean of the Stanford Graduate School of Business, former dean of Harvard’s faculty of arts and sciences and a world-class economist with a John Bates Clark medal and Nobel prize. And last week, Spence was in Washington to make a presentation at a high-powered conference organized by the International Monetary Fund. The moment I heard it, I knew he was on to something important.

Spence and Sandile Hlatshwayo, a new colleague of his at New York University’s Stern School of Business, took the U.S. economy and, from a statistical standpoint, divided it in two.

In one part were jobs in those companies that operate in a global market, subject to competition from foreign firms and suppliers. Most manufacturing wound up in the tradable bucket, but so did agricultural products, minerals, energy and a healthy chunk of business and financial services, from investment banking to management consulting. Not all these companies export, but they are all subject to import competition.

In the non-tradable bucket went activities such as government, health care, retail, construction, hotels, transportation, wholesaling — things that pretty much have to be done here and sold here.

When they looked at what happened between 1990 and 2008 — a period of rapid globalization — two things stood out.

One was that all the job growth came pretty much in the non-tradable activities, in particular government and health care, while across wide swaths of the tradable manufacturing sector, jobs declined significantly.

The other thing they noticed was that in terms of economic value-added — the “output” that is measured by GDP and generally correlates with income — the tradable sector experienced a slight edge.

Put the two together — the very unequal employment growth and nearly-equal output growth — and what you get is an economic tale of two cities, one that is growing in terms of jobs but not income, another that is growing income but not jobs. In short, a recipe for increasing inequality and social and political polarization.

The story behind the numbers goes like this:

Companies in the tradable sector, under the pressure of global competition, are busily outsourcing low- and mid-skilled work to other countries to become more competitive and more productive. The higher-skilled Americans who remain in those firms share in the benefits of that shift through wages and salaries that started higher and have been growing faster.

In the untradable sector, the story is one of rapidly rising employment but not so rapid a rise in output, which has translated into stagnant wages and benefits, both because of the slower growth in productivity and the increased competition in the labor market from all those workers laid off by the tradable sector. Moreover, now that the credit bubble has burst and households and government have had to cut back on their debt-financed consumption, much of the job creation that has gone on in the non-tradable sector is unlikely to continue.

In short, what ails the U.S. economy is primarily a structural problem, not a cyclical one that can be effectively dealt with through the magic of short-term Keynesian stimulus. Unless we find a way to dramatically increase the size and scope of the tradable sector, Spence says, we’re in for an extended period of slow job growth and rising inequality. And make no mistake: at the heart of this problem is globalization.

One response to this dilemma would be to build trade barriers, which Spence, like nearly every other economist, thinks would be a big mistake. But at the same time, he is equally dismissive of free-market cheerleaders who argue that globalization makes everyone better off in the long run — a view, he says, that is not supported by theory or experience. More significantly, it is a view that is increasingly rejected by voters.

Instead, Spence argues that what’s needed is a new policy tradeoff that strikes a better balance between the efficiency and overall economic growth that globalization delivers and the inequality of incomes and job opportunities that it creates.

In Spence’s view, redistributing income from the winners to the losers of globalization can only get you so far. The bigger challenge, he says, is to find a way to tilt the market incentives so that businesses — in particular, big multinational corporations — are more willing to invest in the physical and human capital necessary to make American workers more productive, rather than simply outsourcing work overseas.

We already have a wonderful example of how this might work. In the 1980s, “voluntary” auto-import quotas negotiated with the Japanese led a number of foreign companies to establish assembly plants in the United States. Although the wages paid by the “transplants” weren’t up to the union wages paid by unionized American firms, they were higher than those paid in Asia at the time, and certainly above those in the Southern states where the factories were located. But thanks to extensive training and investments in the latest production machinery, the foreign automakers found they could produce quality cars at competitive prices right here in the United States. Even now, long after the quotas have lapsed, Japanese, German and Korean automakers continue to open plants here. Without that little “nudge” provided by the government, the American auto industry might have gone the way of shoes and textiles.

There is a name for such “nudges” — it’s called industrial policy — and for the past 30 years it has had something of a bad odor in Washington under Democrats as well as Republicans. Larry Summers, until recently President Obama’s top economic adviser, was an unabashed skeptic and Treasury Secretary Tim Geithner seems to be squarely in that camp as well.

There’s no question that sometimes governments get it wrong. There’s also no doubt that governments can get it right, whether it be the lighter touch of European countries such as Germany and Finland or the heavier hand exercised by economic policymakers in China, Korea and Singapore.

What’s pretty clear, however, is that continuing to do what we’ve been doing — pushing down wages and taxes in the hope that free markets will somehow solve this problem — is folly. As Michael Spence and Sandile Hlatshwayo have demonstrated, the future does not lie in further expanding employment in health care, government, restaurants and real estate. Our urgent challenge is to reorient the tradable sector of the economy, services as well as manufacturing, so that it creates not just wealth for investors and jobs for PhDs and MBAs, but jobs for middle-class Americans as well.

The full paper by Michael Spence and Sandile Hlatshwayo, “The Evolving Structure of the American Economy and the Employment Challenge,” will be published soon by the Council of Foreign Relations.