A plain-language companion to the National Balance Sheet, Part 2
The last piece showed America spending down what it owns. This one asks the other half of the question.
For forty-five years the country ran an experiment. Cut taxes at the top, the argument went, and the released capital will go to work — new factories, new jobs, higher wages for everyone. The capital was released. This is about what it built.
The fair answer first
The mechanism is real. Not a myth, not a fraud — it shows up in the data.
The 2017 corporate tax cut is the cleanest test we have, and the definitive study used confidential IRS records to look inside individual firms. Companies that got the average tax reduction raised their domestic investment by about 20 percent compared to companies that got nothing. Across the whole economy, tangible business investment rose roughly 11 percent.
Anyone who tells you tax rates don't affect investment is arguing against the best evidence available. They do.
Now the size of it.
The Council of Economic Advisers promised the corporate cut would raise wages by $4,000 to $9,000 per employee. The measured result was under $1,000. They projected the capital stock would grow 12 to 19 percent. It grew about 7. They said growth would pay for the cut. Corporate tax revenue fell 40 percent, and the extra investment recovered about fifteen cents of every dollar lost.
So: the engine works. It runs at about a fifth of the advertised horsepower, and it does not pay for its own fuel.
Widen the lens and even that mostly disappears. Two researchers built an index of taxes on the rich across 18 wealthy countries from 1965 to 2015, found every major tax cut, and measured what followed. Tax cuts for the rich reliably increased the share of income going to the top one percent. Effect on growth or unemployment: none they could detect.
So where did the money go?
It didn't vanish. Three destinations are documented.
It financed debt. Economists traced top-one-percent savings through the financial system using tax records back to 1963. After 1982, the top one percent's annual savings jumped by nearly 3 percent of national income — over $680 billion a year in today's money. That surge did not fund investment. It funded middle-class borrowing until 2008, and government deficits after that. The rich lent the rest of the country the money the rest of the country no longer earned.
It bought back stock. Share buybacks were illegal market manipulation until 1982, when Reagan's SEC chairman changed the rule. Between 2003 and 2012, S&P 500 companies spent about 54 percent of their profits repurchasing their own shares. A buyback isn't investment; it raises earnings per share by shrinking the number of shares. A regime that released capital on the theory that firms would invest it simultaneously legalized the most efficient possible way of not investing it.
It bid up things that already existed. Buying an asset that already exists is a transfer to whoever sold it, not new capacity. The $200 million painting is the clean case — the artist is dead, the canvas exists, and the country's productive capacity is exactly unchanged by the sale. The same is true, less vividly, of already-issued shares, land, and purely monetary assets.
The gap, and an honest caveat
RAND asked a narrow question: what if incomes below the 90th percentile had kept pace with the economy, the way they did for the three decades after the war? The answer for 2018 alone was $2.5 trillion — about 12 percent of GDP. Cumulatively since 1975: $47 trillion.
Two things need saying. That's a measurement of a gap, not proof of what caused it. Globalization, technology, the collapse of unions, and winner-take-all markets are all candidates alongside tax policy, and anyone waving $47 trillion around as the invoice for Reaganomics is overclaiming.
And here's a complication that cuts against the simple story. Labor's share of national income fell from about two-thirds in 1950 to 58 percent today. But corporate profits captured only 3 of the 8 percentage points labor lost. The rest went to depreciation, interest, proprietors' income, and taxes. "Profits took it from wages" is too tidy. Something broader happened to how output gets divided.
The experiment nobody argued about
Here's the part that should reframe the whole discussion.
In 1960, 94 percent of American doctors and lawyers were white men. By 2010 it was about 62 percent. Talent didn't redistribute itself across the population in fifty years. What changed is that an enormous number of capable people had been prevented from doing the work they were best at.
Four economists measured what removing those barriers was worth. Between 20 and 40 percent of all growth in output per person from 1960 to 2010 came from the improved allocation of talent.
Set that next to the tax argument. Across eighteen countries and fifty years, cutting taxes at the top produced no detectable growth. Over the same period, ending discrimination against women and Black men produced somewhere between a fifth and two-fifths of everything.
One of these was the central economic argument of American politics for forty-five years. The other was filed under civil rights, with an economic footnote.
The bill that came with it
The gain had a cost, and it landed in one place.
Before the barriers fell, American public schools were staffed by women who were capable of far more demanding professional work and were flatly barred from it. The schools got PhD-caliber labor at elementary-teacher wages. That was a vast, invisible subsidy running from women who had no other option straight into the nation's human capital.
When the barriers came down, the subsidy ended. And measured teacher aptitude declined — that's documented.
But the important finding isn't the one usually quoted. The economists who studied it found that compression of teacher pay explains more of the decline than better opportunities outside teaching do. The problem wasn't that talented women left. It was that once they could leave, American schools declined to pay what talent costs — and structured the pay scales so the best teachers gained least by staying.
So: America ran its schools for decades on an off-book subsidy extracted from women with no alternatives. When justice ended the subsidy — correctly, permanently — the country pocketed the liberation gain and never replaced the subsidy with an appropriation.
That's the same story as the first piece, in a different key. An asset was being funded by a hidden transfer. The transfer stopped. The obligation didn't. Nobody wrote down the difference.
The pattern repeats: unpaid care work, still off-book. Foreign-trained skilled workers, whose education America receives and never pays for. Public pensions that keep elders from becoming a direct claim on their children's earnings. Every one of them an input the economy depends on, supplied by someone with limited bargaining power or by an arrangement nobody itemizes — and therefore free to erode with no one recording a loss.
Believers, or cynics?
Both. And we know the proportions because one of the architects said so at the time.
The intellectual argument was real. Robert Mundell, whose work underpinned the policy, won a Nobel Prize. The Laffer curve is trivially true at its ends — a 0 percent rate and a 100 percent rate both raise nothing — so the only question was ever empirical: where's the peak? That's a legitimate question. Treating the whole project as a con is both wrong and tactically stupid, because it concedes that critics aren't serious.
The political packaging was not. David Stockman ran the budget office and was the chief architect of the 1981 cut. In December 1981 he told The Atlantic that the tax cut had always been a Trojan horse to bring down the top rate — that since trickle-down was hard to sell, supply-side was the only way to get a tax policy that was really trickle-down. Confronted publicly, he confirmed it: those were words he spoke.
That isn't a guess about anyone's motives. It's the budget director's own account, published while the doctrine was still being sold to Congress. And the doubt was inside the tent — George H. W. Bush called it voodoo economics in the 1980 primary, months before joining the ticket.
So the honest verdict: a genuine academic argument, knowingly used as wrapping for a distributional goal the wrapping concealed.
Are there still believers, and what's their case?
Yes, and it deserves stating at full strength.
The mechanism is confirmed by the best microdata we have. The counterfactual is unobservable — the 1970s were genuinely bad, and nobody can run the control. Rising inequality doesn't prove growth was forgone; a policy can do both. Capital really is more mobile than it was in 1965, and tax competition is real. And the United States produced most of the world's frontier technology companies over these decades, which any story of simple failure has to explain.
The response isn't that they're wrong about direction. It's about magnitude. Every one of those points is partly right. None of them shows the returns justified the cost — and the best evidence, from researchers with nothing riding on the answer, puts delivery at roughly a fifth of the promise while corporate revenue fell by 40 percent.
Why the next forty years can't look like the last
One question is left hanging by all of this. If median wages barely moved for four decades, how did living standards not collapse?
Four things absorbed the difference.
More hours per household — women entering paid work, with aggregate female participation roughly doubling from about a third in the early 1950s to 60 percent by 2000. Household borrowing — consumption sustained on credit, which is precisely what the top one percent's savings were financing before 2008. Cheaper imported goods, holding down the price of the basket. And falling interest rates, which lifted asset values and lowered the cost of carrying debt.
None of these was illegitimate. Every one of them was a one-time transition, and all four have now run their course. Participation can't double twice. The borrowing ended in 2008. Import prices are going the other way. Rates have reversed.
That matters for reading the productivity numbers. Between 1979 and 2017, economywide productivity rose 68.1 percent while median hourly compensation rose 13 percent. For four decades the gap between what workers produced and what they were paid was papered over by longer hours and more debt. The paper has run out.
So an economy that has been living on the difference between stagnant wages and expanding cushions now has to do something it hasn't had to do since the 1970s: raise the wage itself, or raise what the wage can buy.
The bottom line
Something trickled. Not nothing — but far less than advertised, at a fiscal cost that was never recovered, and mostly into claims on assets that already existed rather than into new capacity.
Meanwhile the biggest measured source of American growth in the whole postwar record was letting capable people do the work they were good at. It required no tax cut. It did have a price — the schools that had been running on those women's underpaid labor were never made whole, and that bill is still outstanding. What it did not require was any of the machinery the country spent forty-five years arguing about.
Full analysis with tables and sources on the National Balance Sheet page, module NBS-2.