Anemic economic growth is the number one issue with the voters. And, as Donald Trump noted in a major policy address in Detroit, “Taxes are one of the biggest differences in this race.” Both candidates favor massive infrastructure spending and grumble about inexpensive imports, but they differ dramatically on taxes.

Hillary Clinton proposes to raise the top tax rate on small businesses to 47.4 percent, to shrink the estate tax exemption by $2 millon, and to impose the highest capital gains taxes in decades.

Imagine you were an economic czar trying to boost incentives for business investment and labor force participation. Would you raise tax rates or lower them? It’s not a trick question, or a hard one. Presidents Kennedy and Reagan answered that question by cutting marginal tax rates on income by 30 percent in 1964–65 and 23 percent in 1983–84. President Clinton cut the capital gains tax by 29 percent in 1997.

Similarly, Donald Trump would “work with” House Republicans’ tax reform plan “using the same brackets they have proposed: 12, 25 and 33 percent.” Capital gains would be taxed at half those rates. Trump would cut the corporate rate more deeply, to 15 percent rather than 20 percent, which could have more bang for very few more bucks.

Unfortunately, partisan critics keep trying to dismiss all such tax reform proposals as “reckless” or “insane” since they promise smaller increases in future revenue than the Congressional Budget Office (CBO) “projects.” Even the status quo flunks that test, however, because – as the graph shows – the CBO baseline projects that revenues from the individual income tax will keep rising faster than the economy forever.

“CBO projects individual income taxes will generate a growing share of revenues over the next decade,” the agency explains, “and by 2025, they will reach 9.5 percent of GDP, well above the historical average.” After that, individual income taxes keep rising without limit — to 9.7 percent in 2028, 10 percent in 2033, 10.4 percent in 2040 and close to 14 percent by 2090. 

These fanciful projections of endless, automatic tax increases, says the CBO, “are mainly because of real bracket creep—the pushing of a growing share of income into higher tax brackets as a result of growth in real (inflation-adjusted) income.”

Since nothing like that ever happened in the past, why believe it will happen in the future?

Individual income taxes averaged 7.7 percent of GDP from 1946 to 2014, and topped 9 percent only five times (1944, 1981 and 1998–2000). The individual income tax brought in just 7.7 percent of GDP from 1951 to 1963 when the top tax rate was 91 percent, and 8.1 percent of GDP from 1988 to 1990 when the top tax rate was 28 percent.

Unlike recent experience, however, the CBO imagines real wages will supposedly rise so rapidly that more and more ordinary people will find themselves shoved up into the top Clinton-Obama tax brackets of 35 percent and 39.6 percent.

The projected future revenues are also “static” which means they assume perpetual tax increases don’t harm economic growth, even though the CBO acknowledges, “Higher marginal tax rates discourage working and saving, which reduces output.”

Every CBO budget estimate warns their “baseline projections are not a forecast of future outcomes.” Yet every attempt to estimate the “cost” of tax reform ignores that warning and misuses these fantastic phantom CBO projections as the standard by which tax reforms are judged.

The Tax Foundation estimates the House Republican tax plan “would reduce federal revenue by $2.4 trillion over the first decades on a static basis.” Due to the larger economy and tax base, however, “the plan would reduce revenue by $191 billion over the first decade.” But note well that such estimates (both static and dynamic) show reduced revenues only in comparison with the rising CBO baseline, not with taxes we actually pay.

Revenues from the individual income tax averaged 8.2 percent of GDP from 2013 to 2015, following Obama’s 2013 tax increase. That 8.2 percent figure is well above any long-term average, partly because of recessions. If receipts remain at that above-average level of 8.2 percent of GDP (which assumes no recessions), then revenues over the next ten years will turn out to be $2.62 trillion smaller than the CBO projected this March.

Keeping individual tax revenues at that relatively high 2013–2015 level (8.2 percent of GDP) would bring in slightly less revenue that over the next ten years than the House Republican plan –even in static terms. And recall that static estimates require pretending (as nobody has) that such dramatic reduction in marginal tax rates on investment, entrepreneurship and education would have zero effect on economic growth.

In short, the House Republican plan is alleged to “lose money” only because it would block “real bracket creep” by repealing the highest tax rates.

Even aside from its vitally invigorating impact on depressed incentives to work and invest, the House Republican tax reform would at most merely thwart a farfetched CBO projection of perpetual tax increases. Relative to recent and historical experience, it would not “cut taxes” at all.