Cross-posted at Reports from the Economic Front.
There are those that argue that lowering the top marginal tax rates on “ordinary” income (from wages or salary) and capital gains will stimulate economic growth. Thomas L. Hungerford, in a Congressional Research Report, tests and rejects this claim.
He finds no statistical relationship between changes in either of these top tax rates and private savings, investment, productivity, or real per capita GDP growth. However, he does find a strong statistical relationship between changes in these tax rates and income inequality. More specifically, raising top tax rates can be expected to promote greater income equality without causing harm to the economy.
Tax Trends
There are two main tax concepts: the marginal tax rate, which is the tax paid on the last dollar of income received, and the average tax rate, which is the proportion of all income that is paid in taxes. How much a person pays on the last dollar received depends on whether it is classified as ordinary income or capital gains.
Most importantly, as the chart below shows, the very top tax payers have enjoyed a steady decline in their average tax rate.
The next chart shows trends in top marginal tax rates on ordinary income and capital gains. The top marginal tax rate on ordinary income has clearly been on the decline: from 91% in the 1950s, 70% in the 1960s and 1970s, to a low of 28% in 1986. It now stands at 35%. The top marginal capital gains tax rate has not changed as much. It was 25% in the 1950s and 1960s, 35% in the 1970s, and is now 15%.
The Tests
Hungerford used econometric methods to test whether changes in top marginal tax rates affect private savings, investment, productivity, and/or per capita GDP growth. Simply plotting the movement of top tax rates and each of these variables suggests that a decline in top tax rates is associated with a positive movement in each of these economic variables.
However, as Hungerford correctly states, correlation is not the same as causation. Using regression analysis, he found that the relationships were only coincidental or spurious; there was no statistically significant connection between changes in the top tax rates and movements in any of the variables.
Hungerford also tested to see if changes in top marginal tax rates had any effect on the distribution of income. The first chart below shows the scatter plot of top tax rates and the share of income going to the top 0.1% for the years 1945-2010. The second shows the same with the top 0.01% of income earners.
As we can see the fitted lines suggest a very strong relationship between the variables. As before, Hungerford used regression analysis to determine whether the relationships were statistically significant. This time his answer was yes in both cases; changes in top marginal tax rates do affect income concentration. In other words, lowering the top rates increases income inequality, raising them reduces it.
It is time for us to start agitating for raising the top tax rates.
Comments 6
Ted_Howard — September 22, 2012
You think this is a serious study? All he did was use a basic time series regression - this garbage would see a desk reject at a top (or really any) economics journal (maybe sociologists standards for statistical analysis is lower).
Let's list the innumerate problems here. Firstly the major one - he assumes all tax changes are exogenous, if they are not exogenous to the economic fundamentals you are trying to measure the results are worthless. Policymakers often change taxes in response to current or forecasted economic conditions, which contaminates simple regressions like the type Mr. Hungerford does - this is why Romer & Romer's narrative approach for identifying when a tax policy shock is exogenous or endogenous has been so influential. Secondly, whether a tax policy change is unanticipated or anticipated matters dramatically for the evolution of economic variables because of expectational affects and intertemporal optimization. For example, standard economic theory predicts that anticipated future tax increases should be contemporaneously expansionary (which puts an interesting twist on Republican claims businesses and consumers are cutting back because of expectations of higher future tax liabilities - optimizing behavior predicts the exact opposite). Thirdly, there is no attempt to control for innovations in monetary policy or government spending, which are often correlated with innovations in tax policy. For example, the ARRA stimulus bill both lowered tax rates and dramatically increased government spending. Mr. Hugerford makes no attempt at any of this, which is why his work belongs in the garbage.
For a serious look at tax policy, try reading Karel Mertens and Morten Ravn's "Empirical Evidence on the Aggregate Effects of Anticipated and Unanticipated U.S. Tax Policy Shocks" :
http://www.arts.cornell.edu/econ/km426/papers/Mertens-Ravn-AEJ-2011.pdf
They find that tax cuts are very expansionary and tax increases are very contractionary. Heck, even research by Obama's own economic advisor (Christina Romer) found the same thing (this paper isn't as thorough, but it's still better than Hugerford's trash):
http://elsa.berkeley.edu/~cromer/RomerDraft307.pdf
Of course, none of this really matters for what we care about - welfare. Redistribution, of course, lowers growth. But what we care about is welfare, not output. Welfare is a much trickier issue to get at and depends on what your welfare function is. As a unabashed supporter of a progressive tax, progressives do themselves no favors when they live in some fantasy world where taxes have no economic effects. We need to make the argument that the declines in growth from higher taxes are sufficiently small, relative to the gains from the recipients of redistribution, that the operation is welfare-enhancing. Of course, I also want a dramatically more efficient tax code, since then we can have more redistribution and a much smaller impact on growth.
Larrycharleswilson — September 23, 2012
Are you suggesting that a neurosurgeon would not do the best he/she was capable of on someone who paid less than another patient?
Vote to Help the Rich, Hurt Yourself « BroadBlogs — November 2, 2012
[...] a low rate on capital gains will encourage investment (a message created in think tanks). There’s no proof of this, but the excuse will do. Fox and friends then spread the word. Now repeat over and over so that [...]
jonalbert — January 24, 2020
Per your health is an online service that provides patients secure access to their health information. Various features may be available on the portal at your practice's discretion, including the ability to send messages to your health care providers, schedule appointments, and pay bills online. Peryourhealth.com