What gains from cap gains?

Justin Fox takes issue with Charlie Gibson:

GIBSON: But history shows that when you drop the capital gains tax, the revenues go up.

I've left out Obama's responses, which were mostly about fairness 'n' stuff, because he failed to give the only appropriate answer, which was that, no, history doesn't show that. Yes, capital gains tax cuts invariably result in a revenue increase the next year, because investors aren't idiots: If they see a cut coming, they're likely to delay capital-gains-generating transactions until after the tax rate drops. But I don't know of any serious economist who thinks that cutting the capital gains tax rate increases revenue over time.

Gosh, I hate to defend either Fox News, or supply-siderism. But this is not the same kind of craziness as claiming that massive marginal income tax cuts raise revenue. Optimal tax theory pretty much hates capital gains taxes because they, as their very name suggests, impede capital formation. Also, capital is much more mobile than labor, which is why countries like Sweden focus their taxation on incomes. In fact, when I look at the graph he posts, it seems to tell me a very different story than it is telling him.

By 2007, capital gains revenues had nearly returned to their 2000 highs in real dollars, even though the indexes hadn't regained their previous (real and/or nominal) heights. When you consider that the capital gains revenues in 2000 were coming off nearly 20 years of uninterrupted growth, this in fact suggests that the capital gains tax raised revenues. Moreover, the inflection point is at the time of the cut to 15%, with revenues marching steadily upward thereafter.

Now, I'd be the last person to suggest that correlation is causation--I'm only pointing out that if they didn't raise revenues, you couldn't prove it by this graph. Moreover, there is a not-ridiculous argument that over the long term--five, ten years--they do raise revenues, by spurring capital formation and economic growth. This is very different from the supply sider argument that you could jam personal income tax rates to 1% and enjoy higher tax revenues therefrom.

Comments

Who makes the supply side argument that you could lower income tax rates to 1% and immediately enjoy higher tax revenues? I believe the real supply side argument is just like the argument you are making here for capital gains tax rate, that at some marginal tax rates you can lower the rate and in the long run gain higher revenues. Of course there is disagreement as to what the optimal rate is, but is is absurd to say that raising rates will always create higher revenues, so it is perfectly logical to say that sometimes lowering rates will increase revenues. Those who say that supply-siders always favor lower rates, down to 1%, are distorting the argument. Read Jude Wanniski, How The World Works.

Posted by Mark | July 10, 2008 4:39 PM

I hate to defend either Fox News, or supply-siderism.

Charlie Gibson is not on Fox News.

The Atlantic's standards are in the toilet, however.

Posted by Jesus H. Christ | July 10, 2008 4:46 PM

Better graph please!

Seriously, what I am supposed to see from that graph? I see that revenues went up, then down, then back up. I see where rates were at only 2 points. Did the rates go up? Because, from only 2 points of information, it looks like rates went down but revenues, after going up, went back down. So based on really crappy information from your graph I'd say that the tax rate makes little difference at all. In fact, the peak looks higher at 20%. So you're saying we should raise rates to the optimal point of 20%? Is that what your graph tells me?

Posted by freddiemac | July 10, 2008 4:50 PM

Perhaps those with questions about the chart should address them to Charlie Fox, since he provided it? Or does nobody rtfa anymore?

Posted by Lorenzo | July 10, 2008 5:03 PM

... Oh please. Nobody claims 1% taxes would maximize tax receipts; this is a strawman so artificial it's embarassing to watch.

(Also, freddiemac: Way to fail at comprehension. All the questions you raised were answered either in the graph, Megan's post, or via a moment's reflection about what other major events other than capital gain tax rate changes might have had an impact.)

Posted by Cody | July 10, 2008 5:38 PM

I agree with freddiemac that you can't get anything out of these data points. All we see is that revenues go up when the economy is good during the tech boom, then they go down after the market crashes, then they start going up again when the economy gets better. Two capital gains tax cuts are stuck on top of this, and there's no effect obviously visible from the chart.

Posted by tennisfan | July 10, 2008 5:39 PM

I read Justin Fox's blog on semi regular basis and he has certainly developed this argument much further in other posts. It essentially amounts to when taxes are cut the troughs and peaks for tax receipts are lower.

As far as this graph, it illustrates his point just fine. In 2001 and 2002 there was a lot of talk of lowering the cap gains rate. The drop in receipts was larger than it would have been, because investors held off selling. The tax cut was made effective and then investors started selling and tax receipts increased. However, since there were tax cuts the peak of receipts in 2007 was lower than the peak in 2000.

Posted by JordanT | July 10, 2008 6:22 PM

Does it make sense that investors would have held off selling in 2001 and 2002 because of talks of lowering the capital gains tax rate? Those were notoriously bad years for the stock market. Selling would have generated a loss and the loss would offset against any gains, which would lower taxable income. Seems like that would be more valuable while the tax rates are higher...

Posted by quick | July 10, 2008 6:34 PM

Megan:

Very big hint: lay the chart above on top of the S&P 500 and see if you notice a correlation.

Posted by mars | July 10, 2008 6:35 PM

Not surprisingly, when you create a category of income that's taxed at a much lower rate you create an incentive for people who have some flexibility in their compensation to move as much of that compensation as they can from the higher-taxed category to the lower-taxed category. Therefore, revenues for the lower-taxed category will increase. Certain hedge fund managers, for example, have managed to have their commissions categorized as capital gains. CEOs and other top executives have also moved much of their compensation into the capital gains category.

Why this should be considered a good thing or something we should encourage is beyond me.

I find the entire premise flawed. Why do we tax income made from investments at a much lower rate than money made from work? I understand that we'd like to encourage investing, but shouldn't we also encourage working? Isn't work a good thing too?

Posted by Chuchundra | July 10, 2008 6:44 PM

On the broad issue of "why a lower tax rate for capital gains?" the answer is (as I understand it) the time value of money, and to some extent to offset risk.

If I have income from working, I get it when I do the work (give or take a month). When I make an investment, I agree to tie up my money for a substantial period of time, in the uncertain hope of an eventual profit. If, say after five years, I realize my gain, I've foregone the use of that money for five years, as well as placing it at risk of loss. (Yes, the ability to use losses to offset gains helps some, but only if I make enough gains to balance my losses.)

Since the cap-gain tax advantage is a recurring issue in this thread, perhaps Megan will address it?

Posted by Lorenzo | July 10, 2008 7:17 PM

What I see is that the receipts were in creasing from 95 to 97 at the same rate as they were after the rate was dropped. Overall, I am disappointed as a regular reader but rare commenter in your analysis here. Two data points on an up and down graph, not comparing it with the overall economic picture or at least the DOW? Weak.

Steve

Posted by steve | July 10, 2008 8:15 PM

Megan,

Include a graph of GDP along with your cap gains graph. It'll show when the economy is growing and when it's in recession. You should also include a graph of the S&P500, so you see how the market is performing.

BTW, the CBO published a very detailed analysis of who pays taxes last winter. It shows each of the quintiles, plus the Top 10% and Top 1% over that last 30 years (I think - forget when the plot starts).

Back to the 3 plots on one graph. Once you've done this, you're analysis of cap gain tax and its revenue creation may be a little more nuanced!

Posted by Mark Michael | July 10, 2008 9:31 PM

Y'all all completely missed the point. Fox is arguing that the cap gains revenue goes up randomly; in fact, this graph shows inflection points at each drop. As I explicitly said, I'm not arguing that the graph proves that cap gains cuts raise revenue; only that the graph does not seem to prove what Mr. Fox is arguing. If anything, it is vaguely suggestive of the opposite of what he is saying.

Posted by Megan McArdle | July 10, 2008 11:15 PM

Megan, as other commenters have pointed out, your "you could jam personal income tax rates to 1% and enjoy higher tax revenues therefrom" is a serious straw man attack.

Just about no one would actually make such an argument.

With high enough tax rates cutting the top marginal rate will increase revenue even in the short run. At some lower point cutting the top marginal rate will increase revenue in the long run. Neither of these points are anywhere near 1%. Where they are exactly is uncertain, controversial, and something that changes over time, but the supply side argument is NOT "all tax cuts increase tax revenue".

Posted by Tim Fowler | July 10, 2008 11:56 PM

"Y'all all"? It "alls of y'all". Sheesh.

Posted by mr. snrub | July 11, 2008 12:12 AM

It's. Doh.

Posted by mr. snrub | July 11, 2008 12:14 AM

Fox is arguing that the cap gains revenue goes up randomly

He said nothing about randomly, just that the changes to the capital gains tax rate affects the amount of tax received only in the short term. In the long term, at our current tax rates, a drop in our capital gains tax results in a drop in receipts.

Since the cap-gain tax advantage is a recurring issue in this thread, perhaps Megan will address it?

The advantage is that the rich and powerful are able to direct their income into "capital gains" If investing and having your money illiquid makes for a special tax rate, then why does CD interest count as income?

Posted by JordanT | July 11, 2008 1:07 AM

Lower capital gains tax must mean less revenue in the case of people who can structure their income to become capital gains...Warren Buffet for example.

Posted by Aaron | July 11, 2008 5:33 AM

Hm, an inflection point is a point of zero curvature (f''(x)=0), not a point of zero slope (f'(x)=0). Unless this is some informal usage of the term, which I heartily disapprove of anyway: why muddy a perfectly precise notion?

Posted by hw | July 11, 2008 12:31 PM

_Why do we tax income made from investments at a much lower rate than money made from work?_

I'm puzzled about the same thing. Suppose I buy a bunch of mopeds that I rent out to tourists in the summer. My profits are taxed at the income rate. But then suppose you skip the mopeds part and just rent out your money--you make loans and collect the principle and interest at a later date. Your profits are taxed at the lower capital gains rate. But why is being a banker more special than a moped renter? If the answer is, "it's not", then basic fairness requires the tax rate to be the same.

But maybe there is something special. Renting moped and money are both risky ventures, so that's not it. Is there some sort of productive leverage gained from my being a banker instead of going into another line of business that benefits us all?

I can see that this lowers the interest rate the investor needs to make it worth his while. But why not let the market determine the price of trading consumption and leisure (what the investor and borrower pass up) for investment and production (which pays off for both down the road), undistorted by the subsidy?

Posted by DKE | July 11, 2008 4:33 PM

DKE - If you loan money and get interest you don't pay the capital gains rate.

Posted by Tim Fowler | July 11, 2008 5:09 PM

The whole premise is ridiculous.

It is rather obvious that if "capital gains tax rates" are lower you will generate more CAPITAL GAINS TAX revenue, as more individuals with the means will move more money into these instruments.

But, you will have sacrificed INCOME TAX revenue which is at a higher rate which means a lower overall tax revenue for the government in the short term, but with (hopefully) more general economic growth through investment.

jeez.

Posted by nbchi | July 11, 2008 5:09 PM

Tim: Thanks for the correction. I'd meant the lending thing to stand in for capital investment in general, but I can see that it muddies the waters. Anyway, make my example clearer by supposing that instead of lending to a business, you buy stock in it. Then ask all the same questions. What's special about buying stock as opposed to renting out mopeds (or loaning money!).

Posted by DKE | July 11, 2008 5:39 PM

Steve, good point, as is another previous one that Megan has shown convincingly that capital gains revenue increases when the stock market increases.

But also note that the rate of increase in 95-97 appears to be about the same after the rate was cut, even though the market moved up even faster into bubble conditions in the late 1990s.

My other favorite bad logic in this area is how people try to show that capital gains dropped when the rate increased in 1986. Of course, the rate increase was coming so people sold out just before the increase.

Posted by mike | July 11, 2008 9:42 PM

I do not believe a supply-sider would say that a 1% rate would increase revenue. The concept, which is undeniable in the abstract, is that at 0% rate, you get 0 tax revenue, and at 100% rate, you kill the economy and get 0 tax revenue. Therefore, there must be a situation where an increase reduces revenue and a decreases increases tax revenue.

However, and while I realize the top rate starts at different amounts and the 1986 law changed passive loss rules, we have had good economies at 90% (1950s), 70% (1960s), 50% (1980s), 39% (1990s), 36%, and lower rates. I suspect that the magic point where decreases in tax rate increase revenue is WAY higher than current rates.

Obviously the tax increases under GHWBush and Clinton did not kill the economy, and at the end of the run, we had huge surplusses.

Posted by mike | July 11, 2008 9:53 PM

You get paid for your work on an ongoing basis.

You collect interest, normally, on an ongoing basis.

In capital gains you put down a bunch of money up front and don't collect anything in return for perhaps several years. Your initial investment is affected by the intervening inflation - AND your first-year presumed profit is affected by the intervening inflation after the first year - AND your second-year presumed profit is affected by the intervening inflation after the second year - etc.

Many people have suggested that we could comfortably give up the lower tax rate for capital gains, in exchange for adjusting the basis (amount invested) for inflation. But the government will never go along with this because it'll point out to people how much the government is inflating away our money.

(So why the special rate for dividends? A different argument. Wages and salaries, normally only the recipient pays income tax on. Interest, normally only the recipient pays income tax on. Dividends, both the payer and the recipient pay income tax.)

Oh, and Mike, Clinton ran such huge "surpluses" that the national debt went up every year, and that's not including the unfunded liabilities of Social Security and Medicare. This while benefiting from a huge stock-market bubble plus the apparent general prosperity needed to give people the cash to pour into that bubble. (Granted the current spendthrift-in-chief is even worse in terms of financial restraint.)

Posted by warrl | July 13, 2008 3:44 AM

Thanks warrl. If you're still checking in on this thread, a follow-up for you.

The principled case for lower cap gains tax is the effect of inflation on your investment base, and on presumed profits. Now my question: assuming a stable business, won't there be an increase in the value of the investment base and profits that tracks the inflation rate?

Suppose the company I bought stock in sells bread. The value of the business would go up as the price of bread goes up. That would be reflected in an increased share price. And their profits (in absolute dollars) would go up as the price of bread goes up (assuming a constant profit rate). When they plow these profits back into the business, this is also reflected in an increased share price. If this is right, then wouldn't it also be true that my investment base, and the profits from that investment, will increase along with the rate of inflation?

Based on the above, it seems that the higher return required to incentivize stock investment has very little to do with the effect of inflation and much more to do with the uncertain performance of the business.

Posted by DKE | July 14, 2008 11:04 AM

Wait, I think I got things backward. The argument is that cap gains *do* reflect inflation-based increases, with the problem being that taxes are assessed to nominal gains, not real gains. So, if gains were indexed to inflation, then there'd be no principled objection to taxing them at the regular income tax rate. Is that right? If so, fine by me, let's index. (While we're at it, let's get rid of the corporate tax and hit dividends with the same income tax rate.)

Posted by DKE | July 14, 2008 11:25 AM

DKE - As warrl pointed out a lot of many capital gains is nominal, not real. Also using the specific example you gave me - "make my example clearer by supposing that instead of lending to a business, you buy stock in it" - Well if you own stock your a part owner of the company, and if the company is making profits, its probably paying taxes on them. Your property is already paying income taxes. That could argue for you not paying any taxes on the capital gain, except that you can make a capital gain on stock of a company that doesn't pay corporate income tax, or on things besides stock (say a house, a painting, a rare old car, etc.).

It might make sense to move the rate for capital gains up to the ordinary income tax rate (or better yet move that rate down to a point where it doesn't exceed the capital gains rate), while at the same time indexing capital gains to inflation AND getting rid of the corporate income tax.

Posted by Tim Fowler | July 14, 2008 1:06 PM

mike - re: "My other favorite bad logic in this area is how people try to show that capital gains dropped when the rate increased in 1986. Of course, the rate increase was coming so people sold out just before the increase. "

That's true, but it doesn't imply that the higher capital gains rate doesn't 1 - Decrease the incentive for investment and thus put downward pressure on investment (investment may still go up if other factors put a larger amount of upward pressure on investment), and 2 - Increase the tendency for people to hold on to their investments when they would have otherwise sold, hoping to defer paying the higher rate (or even never pay the higher rate if they hold on until the rate goes back down, since rates do fluctuate).

The first response means less investment than there otherwise would have been. The 2nd response most likely results in slightly less efficient investment that there would otherwise have been.

(Note I'm talking about both effects occurring at the margin, its not like its a night or day thing , raise the capital gains rate by 5% and its not like most people would hold on for years when they otherwise would have sold)

Posted by Tim Fowler | July 14, 2008 1:21 PM

mike - re: "However, and while I realize the top rate starts at different amounts and the 1986 law changed passive loss rules, we have had good economies at 90% (1950s), 70% (1960s), 50% (1980s), 39% (1990s), 36%, and lower rates."

As you yourself point out the level when the top rate kicks in has varied a lot. Also so has the number of exemptions, deductions, and loopholes. Almost no one paid 90% or even 70% in federal income tax. Not only was the amount you would have to earn to fall in to that rate very high, but normally you could find all sorts of ways to avoid paying that much even if your income was that high, and you certainly had lots of incentive to do so.

Re:"Obviously the tax increases under GHWBush and Clinton did not kill the economy"

It takes a lot to kill the economy. There are always other factors in play, the current top personal income tax rate is just one of them. Which doesn't mean that the increases did not harm the economy, but "harm" doesn't equate with "kill", and while at the margin the higher rates where harmful, they where not the only thing going on in the economy. Just because you add one harmful effect doesn't automatically keep you from having a good or even a great period for economic growth.

Posted by Tim Fowler | July 14, 2008 1:28 PM