After laying out a very broad economic agenda in early July, Hillary Clinton returned last week to start drilling into specifics.

These include combating what Clinton terms "quarterly capitalism" — the tendency of corporations and businesses to focus on short-term gains to shareholders, rather than long-term investments in jobs, better pay, and more productive capital.

Unfortunately, the way Clinton frames the problem has a self-defeating quality to it — one that hamstrings her diagnosis and proposed solutions right out of the gate.

To be sure, Clinton is on the right track. Now that the early-'80s revolution in stock buybacks has been piled atop traditional dividend payments, profits have been almost completely swallowed by the money companies pump out to shareholders. (In fact, it now looks like equity — the money representing shareholders' ownership stake in companies — is a net drain on corporate assets, and companies are literally borrowing to increase the flow of cash to investors.)

"Large public companies now return $8 or $9 out of every $10 they earn directly back to shareholders, either in the form of dividends or stock buybacks, which can, temporarily boost share prices," Clinton said. "That doesn't leave much money to build a new factory or a research lab, or to train workers, or to give them a raise." Profits and net investments have broken free from one another in recent decades, with both scientific research and overall investment on the downswing. And of course, wages and compensation for middle- to lower-class workers have stagnated or declined.

The problem comes in when Clinton tries to frame the fight against "quarterly capitalism" as a kind of win-win for everyone. She argued we need to re-conceptualize what we mean by shareholder value — "long-term growth, not short-term profits," "building companies, not stripping them," etc — and that doing so is how we "raise incomes and deliver real value for shareholders."

But any shareholder looking at these trends could be forgiven for thinking the value they're getting already is quite real enough. Since 1996, dividends, stock buybacks, and other forms of capital gains have contributed more to rising inequality than any other form of income. Nor is there any evidence this flow of money is unstable over the long term: Both the stock market and high-end luxury industries recovered quickly and handily from the Great Recession, and corporate profits are at a 60-year high as a share of the economy.

To be frank, what Clinton needs here is a hefty dose of traditional class war.

What we're talking about is the flow of revenue within firms. Does it go to new hires, higher pay, and investments in productive physical capital? Or does it go to shareholders and massively over-compensated CEOs? One thing the money can't do is go both places at once. Over the last few decades, the money hose going to the latter groups has increased enormously. There's nothing the hose going to the rest can do except shrink in response.

That clarifies what's inadequate about Clinton's central policy proposal on this subject. Right now, when someone sells stock they have held for less than a year, then the money they get is taxed as a short-term capital gain, which means it’s taxed as ordinary income. If they held it for more than a year, it's taxed at the extra low rates for long-term capital gains. Clinton wants to stretch that to at least two years, and introduce a six-year sliding scale for earners in the top tax bracket.

That's certainly not a bad idea. But it's unclear how much punch it would really have. Tax rates on long-term capital gains have been all over the place over the last century, and there's little to no evidence they had any effect for good or ill on long-run investments or economic growth.

Clinton's focus on the timing issue — how long people keep their money parked in one company — arguably gets things backwards. It's possible to whip short-term thinking without doing anything about the size of the money hose to shareholders. Conversely, if you shrink that money hose, does short-termism really matter anymore?

This is the problem with saying "more economic growth" when what you really mean is "more jobs." The two are not necessarily the same thing. Our economy is structured to take enormous portions of the wealth it generates in any particular year and funnel it to the top. Obviously, as long as that's the case, wage stagnation for the middle and bottom is going to be an ongoing problem. And if you're resigned to your economy being structured that way, then yes — endless pursuit of high growth rates is the only way to combat wage stagnation.

But the much simpler solution — and the one that can ensure broadly shared prosperity even when long-term economic growth rates are low — is don't structure your economy that way. Redesign the annual flow of money generated by economic growth so that it favors the wealthy less, and the middle and the poor more.

If Clinton wants to restructure that flow, she needs to remember what matters: increasing workers' bargaining power to claim a bigger share of their company's revenue. Hiking the minimum wage would be a good start, and to her credit Clinton supports an increase of an as-yet unspecified amount. Another thing Clinton has left the door open for, but hasn't given details on, is reforms to regulations of corporate governance. Altering the law to halt or roll back the decline of unions would be another good move, as would reorienting the Federal Reserve to prioritize full employment above all.

Finally, what higher corporate and capital gains taxes did mid-century was make it exceedingly expensive to take out revenue as profits and shareholder payouts. They massively shrunk the size of the shareholder money hose, rather than just tweaking the timing of the long-term capital gains definition. They didn't ask shareholders to re-conceptualize "real value." They operated on the presumption that companies should be driven by the tax code to pump as much money as possible into investments and jobs — whether shareholders liked it or not.