It was an awkward moment for Gary Cohn, the former top Goldman Sachs executive who is now Donald Trump’s lead economic adviser.

Cohn was making an appearance at the Wall Street Journal‘s mid-November CEO Council conference. Associate editor John Bussey asked the audience packed with corporate executives for a show of hands about how many planned to increase their company’s investment if the Trump administration’s tax reform bill becomes law. Very few hands went up.

“Why aren’t the other hands up?” a visibly abashed Cohn asked.

What makes this so awkward is not what it tells us about Trump’s tax plan but what it says about Cohn’s lack of understanding of how the tax plan works. Cohn’s bewildered response appeared to bolster the arguments of critics who say that corporate tax cuts will not spur the economy to accelerate.

Critics of the proposed corporate tax cuts basically think that if they worked, they would work according to a “trickle down economics” strawman: cut taxes for business and they’ll somehow pay the new profits to the workers. Then they disprove this strawman by arguing that businesses will not use their additional after-tax profits or money repatriated from foreign subsidiaries to create jobs, raise wages, or invest domestically. Instead, the corporations will send the gains to shareholders in the form of dividends or share buybacks.

“Major companies including Cisco Systems Inc., Pfizer Inc. and Coca-Cola Co. say they’ll turn over most gains from proposed corporate tax cuts to their shareholders, undercutting President Donald Trump’s promise that his plan will create jobs and boost wages for the middle class,” Bloomberg News reported.

This is a misconception about how corporate tax cuts work that is rooted in criticism originally directed at the much-maligned and little-understand repatriation tax holiday of the Bush administration. The “Homeland Investment Act” of 2004 gave international companies a large one-time tax break on overseas profits, but only if the money was used in ways specified by Congress. Hiring, debt-restructuring, and R&D were allowed, but dividends and share buybacks were not.

Studies on corporate behavior showed what everybody should have known: the prohibitions on buybacks and dividends were largely ineffective because money is fungible. Companies could use repatriated foreign profits to fund the permissible activities that they would have funded with domestic profits while using those domestic profits to increase dividends and buybacks. U.S. companies brought back around $312 million and an estimated 92 percent of it went to shareholders.

Why would companies do this? Why not spend to expand their business and hire new workers? The primary reason is that the companies with the largest amounts of cash sitting abroad were not starved for capital. They did not need to repatriate any funds in order to fund a planned business expansion, since they had plenty of domestic profits and could borrow cheaply.

As the authors of the ground-breaking study on the 2004 repatriation holiday “Watch What I Do, Not What I Say: The Unintended Consequences of the Homeland Investment Act” wrote: “If firms are not financially constrained, the well-governed firms would return any internal capital accessed under the HIA to shareholders through mechanisms such as share repurchases or dividend payments.”

In other words, the firms with the biggest cash hordes abroad did not need to repatriate that cash to invest in their business. So instead, they paid the cash to shareholders.

This behavior of the repatriators, however, has led a lot of journalists and economists to assume that this meant repatriated dollars did not go to support job growth in the U.S. It’s as if they assume that money paid to shareholders simply gets extinguished from the economy rather than recycled as additional investment or consumption.

The authors of the repatriation study knew better:

Although the H.I.A. does not appear to have spurred the domestic investment and employment of firms that used the tax holiday to repatriate earnings from abroad, it may still have benefited the U.S. economy in other ways. The tax holiday encouraged U.S. multinationals to repatriate roughly $300 billion of foreign earnings and pay most of these earnings to shareholders. Presumably the shareholders either reinvested these funds or used them for consumption. Either of these activites could have an effect on U.S. growth, investment and employment.

In other words, you should not survey corporate leaders at the biggest companies on what they will do with additional earnings from tax cuts or repatriated profits because these are not capital constrained. For the most part, they can access all the capital they need to expand their businesses as desired. If they’re holding additional cash abroad or not expanding, it’s because they do not see appropriate opportunities to spend on hiring new workers or investing in R&D.

To the extent that taxes are driving this, it means that the tax code is trapping funds inside of large corporations that are not using it productively. On a large enough scale, this can slow economic growth and depress wages. A corporate tax cut–especially one that encourages the repatriation of foreign earnings–frees this cash up to be spent and investors by others outside of the purposes of corporate America’s leaders.

Dividends and buybacks are not a sign that a corporate tax cut is not working. They’re a sign that it is working to redistribute American capital away from the biggest and most successful companies that do not need it and toward businesses that do.

This is not just some theoretical idea about how buybacks and dividends can improve economic performance. Although it can be harder to measure, there is good evidence that distributing cash to shareholders and getting it out of corporate vaults does increase employment. A recent study took advantage of what’s known as “home bias” in investing to figure out how buyback driven by the 2004 repatriation holiday indirectly contributed to job creation.

“Home bias” is the well known if somewhat mysterious fact that shareholders of public companies tend to live in the geographic region around the company’s headquarters. We do not entirely understand the reason for this “home bias” in investors but it probably has something to do with individual investors preferring shares of companies they or people they know worked for and fund managers preferring companies whose executives they have access too locally. Because of home bias, economists can study the effects of dividends or buybacks by studying the economic performance of the neighborhood around a corporate headquarters.

The study found that the areas around companies that repatriated funds and distributed it to shareholders had more job growth than similar areas whose local businesses didn’t repatriate funds. In fact, there was a positive relationship between how much was distributed to shareholders and how well the local job market did. The authors of the study, Duke University professors Scott Dyreng and Robert Hills, interpret this as evidence that the repatriation holiday “increased domestic employment, consistent with legislative intent.”

It’s important to keep in mind that the alternative to tax reform is not somehow forcing the CEOs to hire or invest more. It’s to not pass reform and encourage them to keep hording cash. Why would progressives or populists ever advocate for a system that keeps piling up cash in the vaults of America’s CEOs?

In other words, corporate tax reform that distributes money away from corporate managers who do not use it to create jobs or grow their business and to shareholders who can use the money to increase consumption or invest in growing but capital-constrained businesses is an economic populist reform. It is democratizing control of profits away from CEOs of America’s largest companies.

Too bad that even some of Trump’s top people do not understand how it works.