The new Brookings Institution study says that income polarization isn’t as bad as we’ve been told. A new profile of “reshoring” manufacturing jobs argues that it’s China that’s losing jobs, not America. We’re still down seven or eight million manufacturing jobs since our peak in 1979, but ignore that, so the reasoning goes—we’re on an uptick, with rising wages down South. The Obama recovery is real, with 14 million new jobs over the course of his presidency. We should be talking about what’s getting better.

Structural changes in the American economy are difficult for our political system to digest. And regional patterns make it hard to speak credibly. But the Rust Belters are not crazy to vote, massively, for the outsiders Sanders and Trump. There are indeed millions of former manufacturing workers whose lifespans are shorter and whose stress levels are higher, as counted by no less credible a scholar than 2015 Nobel Economics Prize-winner Angus Deaton, and they are indeed concentrated in those parts of the country where deindustrialization—and globalization—have not been softened by brand-new, incremental growth in jobs that pay around twice the federal minimum wage.

We so, so want to be positive. Yet even in the new manufacturing jobs there’s a historic change: Labor as a factor of production has radically shrunk. Take the case of Erie, Pennsylvania, just 80 miles from here, which is losing its General Electric locomotive factory. The 1,500 union workers who were sent layoff notices in Erie in 2015 compare to 500 non-union workers now employed in Forth Worth, Texas. There are many more Erie layoffs to come; the assembly work in both locomotives and in propulsion systems for huge off-highway vehicles, like those used in the Alberta tar sands, is already gone. Soon, the 1,500 salaried workers—including lots of engineers—will dwindle in Erie as the manufacturing and assembly departs. The company argued that it wasn’t breaking any contract with the union because GE Manufacturing Solutions is a different entity than GE Transportation, and that therefore it wasn’t union-busting to move. Labor is about seven percent of the cost of producing a locomotive. Moving from Erie to Texas was not about money, insiders say: it was about who had the power.

Thus when Sanders decries the “rigged economy,” and when Trump yells about corporations that ship American jobs away, there are tens of millions of Americans who have direct, personal experience that somebody famous is validating. Recently, the Carrier air-conditioning company told its workers that their jobs are moving to Mexico. So candidates traversing Ohio, Illinois, and Missouri today ask for support from people whose skills, whose workplaces, and whose communities have been deemed no longer good enough. Was it they who decided to commit collective economic suicide, or was it the corporate elite whose salaries are 300 times what an Erie union job pays—excuse me, paid?

And then there’s the memories of some northern labor people who talk to Arkansas labor people about Hillary Clinton, who served the Walton family enterprise called Wal-Mart as a board member for six years. President Bill Clinton negotiated the North American Free Trade Agreement, which some in labor have angrily mythologized as the only reason for the loss of good-paying jobs. We’re talking politics now: No political analyst should expect that a mainstream candidate like Clinton will get the kind of support that Sanders and Trump will get from people with personal experience of lob loss, wage loss, and family breakup as a result of capital flight.

Political numbers?

The bloodless assertions of those economists writing Brookings’ study of incomes and wealth are like the glib generalizations of the apologists for the big jump in manufacturing jobs in Texas. The economists do not dispute the general findings of Thomas Piketty and Emmanuel Saez about incomes jumping at the top and lying flat at the middle, but parsing the data these new guys are using raises the eyebrow.

Saez’s recently efreshed data that show how great the skewing of income has been toward the top one-tenth of one percent of households in the US; Saez uses actual IRS returns to report income, while the Brookings people rely on the self-reporting of high-income people participating in surveys, and on some estimates of the expected lifetime value of Social Security and of some private pension payments. (One prefers the IRS data, because it’s actual income by real individuals, not educated guesstimates.)

George Orwell essayed famously about politics and language. Today, our data have been politicized. But the experiences, the countable, measurable experiences of people in the Rust Belt do not yield to these new numbers.

We’ve long had political hacks like Martin Feldstein, the ever-available Harvard economist, who served Nixon, Ford, Reagan, and the presidents Bush, and who can be counted on to pooh-pooh income polarization when he’s not busy ducking questions about his membership of the board of directors of AIG insurance, the prime marketer of egregiously fraudulent credit-default swaps and other derivatives that led the crash of our economy in 2008.

But these days, the rise of Bernie Sanders has recently turned even Keynesian economists into Feldsteins. Genuinely smart and well-regarded people like Paul Krugman and Paul Starr sound like old-time anti-communists, in that they are very loose with very old language.

Krugman, long a very firm and very widely published advocate of single-payer universal healthcare (see his long series in the New York Review of Books in 2006 and 2007) jettisoned his former views when Bernie Sanders endorsed “Medicare for all.” Krugman wrote in 2011 about the “game change” moment that had just happened, when American imports from Third World economies outstripped trade with peer democracies. It was Krugman who radically criticized trade agreements, including NAFTA and deals with China, because they were bad for American workers. Paul Starr, also a Princeton professor, also known for expertise on healthcare, now publishes slashing attacks on Sanders. Of course they dump on Trump, too—but these guys are bashing Sanders for quoting them!

But the numbers is the numbers.

First, the wreckage that globalization left behind in the Great Lakes region is not being repaired by the brand-new manufacturing jobs coming back to some parts of America. In these pages, you have seen some of our ongoing metro-by-metro survey of where the African-American men are not only out of jobs but entirely out of the workforce, like here in Buffalo-Niagara Falls, where only 48 percent of black males aged 20-64 are in the workforce. This is precisely the experience of black men in Milwaukee, Chicago, Detroit, Toledo, Cleveland, etc.—and our data, real noses counted, not estimated, by the local offices of State labor departments, show this: that the former manufacturing centers of the USA are precisely where the manufacturing jobs aren’t coming back.

It’s also in these metros where there’s sprawl without population growth, where the new jobs in the Obama Recovery are real and they are welcome, but where they pay minimum wage and don’t come with benefits. These are the places where, thanks to Donald Trump’s campaign, the politics is getting Redder than even the Republican governors of Wisconsin, Illinois, Indiana, Ohio, and Michigan have already made it.

It will be no surprise if the winners of the primary elections in Illinois and Ohio will be the two outsiders—especially in Illinois, where Hillary Clinton’s old family retainer Rahm Emmanuel is not only deeply unpopular, but where Rahm’s friend Governor Bruce Rauner just widened the racial and class divide by shuttering majority-black Chicago State University and laying off all its professors, administrators, and support staff only a few days before Donald Trump’s partisans tussled with protesters on the U of Illinois’s downtown campus.

Chicago has been called one-third San Francisco and two-thirds Detroit. But Chicago and Detroit are renaissance paradises compared to the rest of the Rust Belt towns, like St Louis, a sprawled-out, shrinking urb that just lost its football franchise, where there’s a suburb called Ferguson and a two-generation-old economic and social disaster called East Saint Louis just across the Mississippi, and the same kind of numbers we see in Buffalo.

Promises, premises

European Social Democrats have been waging electoral wars against austerity-minded, banker-influenced Conservative parties in Greece, Spain, and Portugal. The Left there has gotten traction because youth unemployment is insanely high, and because the bankers are unpopular. In very broad strokes, similar sentiments have found political voice here, as both Sanders and Trump have complained about the infrastructure deficit, have pledged to have something more comprehensive than Obamacare, and in Sanders’s case, have been very forceful about raising income taxes on financial speculators (Trump has been explicit only about eliminating the low tax rate on hedge-fund managers and on people who, like Mitt Romney, enjoy the “carried interest” loophole.

Sane, non-partisan people know that the federal budget deficit has shrunk under Obama, but that there’s absolutely no prospect of raising tax rates on the tip-top-income households unless a Democratic majority in the Senate and in the House support a strongly progressive Democrat.

And Sanders, though he is not favored to win the nomination, is observed to be driving turnout among working-class whites, Millennials, and independent voters of every demographic group. So the only chance for a Democratic tax policy ever being enacted probably rests with Sanders as the Democratic presidential nominee.

Smart people with long track records as labor partisans have endorsed the Sanders tax and healthcare plans, and most have been unlike Paul Krugman and Paul Starr: most still endorse tax progressivity and universal Medicare despite there now being an actual, viable presidential candidate Smart people with long track records as Republican partisans have decried Sanders’s proposals, but have not brought themselves to endorse Trump’s non-specific slogans.

But these are academic exercises, and will remain so, until such time as Sanders becomes the Democratic nominee.

Should Clinton prevail in the delegate-selection race, the prospects of even holding onto Obama’s incremental healthcare and tax reforms will be slim, given that her candidacy is not expected to drive the turnout that the down-ticket Democratic candidates will need in order to prevail in Trump-dominated media markets.

So it’s looking like this: the South, where Democrats lose electoral votes and congressional races, will continue to benefit from the return of manufacturing to the USA. The North, where Democratic voters are energized by Sanders, will probably go for Trump—and a Republican Congress, and Republican state legislatures—if the Democrat Trump faces is Hillary Rodham Clinton.

So without a Sanders victory that sweeps a 2008 or 1964 or 1932-style Democratic House and Senate majority in, don’t expect anything to change in the profile of who wins in the new economy.

And if nothing changes in the new economy, we have a bitter future ahead—a future of masses of alienated, economically obsolete Americans with no real shot at the prosperity they can see on TV. That’s to say, America will look like the Rust Belt.

Bruce Fisher is visiting professor at SUNY Buffalo State and director of the Center for Economic and Policy Studies.