Recent research released by prominent economists Peter Diamond and Emmanuel Saez argued that taxing top earners at an even higher rate of 73% produces the maximum government revenue. The study ignited debates among economists and policymakers alike.
However, economists Alejandro Badel and Mark Huggett wanted to take a deeper look at taxation by using numerical methods to simulate a complex economy, which led to their collaborative research and subsequent working paper, Taxing Top Earners: A Human Capital Perspective.
The economists' initial findings yielded vastly different results than those of Diamond and Saez, including a lower optimal tax rate and the added consideration of human capital accumulation.
A Model Economy
Creating a model economy was the first step in evaluating the consequences of increasing the tax rates on top earners. The artificial economy is calibrated to match the U.S. earnings distribution, where earnings and wealth differ vastly across the population.
"When one builds an economic model, a typical assumption is that agents make best choices to maximize their welfare," said Huggett an economics researcher at Georgetown University. "Solving a maximization problem by computer can take a particularly long time if the number of possible states of individuals are very large."
"We need resources like Stampede to efficiently compute optimal individual decision rules about working, learning and saving, and then simulate the behavior of millions of model households, using these rules, in order to determine economic aggregates and equilibrium behavior," Badel said.To address this computational challenge, Badel and Huggett turned to XSEDE, the Extreme Science and Engineering Discovery Environment, which provides researchers access to shared computing resources, data, and expertise. Through XSEDE, the researchers ran simulations of their model economy on Stampede, one of the world's most powerful supercomputers, at the Texas Advanced Computing Center (TACC).
"We need resources like Stampede to efficiently compute optimal individual decision rules about working, learning and saving, and then simulate the behavior of millions of model households, using these rules, in order to determine economic aggregates and equilibrium behavior," Badel said.
Initial simulations of the model economy revealed 53% as the optimal tax rate for top earners, much lower than the 73% rate Diamond and Saez proposed.
Human capital accumulation also plays a large role in the economists' model economy, affecting both the shape of the Laffer curve and optimal tax rate. In the model, top earners have a combination of good initial conditions, which include a high skill level and high learning ability by age 23. They also invest heavily in acquiring new skills, even if that means accepting reduced earnings.
But the researchers argue that an increase in the tax rate would have adverse affects on these potential high earners.
"Suppose you're a 25-year old in medical school, and considering taking on a medical specialty that might involve many years of hard work and lower earnings. Yet, you also know that once you become a great doctor, you will be compensated very well," Huggett said. "If you learn that the top tax rate increased from 42.5% to 73%, you may decide taking on that medical specialty is not a good decision."
Badel and Huggett found that a spike in the tax rate would have adverse affects on potential high earners.Badel and Huggett argue that these potential high earners would be most strongly impacted by increasing the marginal tax rate. Higher taxes would lead to a change in skill-related investments for this group, resulting in a reduction in human capital accumulation and eventually reduced government revenue.
A New Perspective
Even though Badel and Huggett are still running simulations of their model economy to better determine the optimal tax rate for top earners, their research has already garnered attention in economic circles.
The two economists have also been disseminating their initial research in the U.S. and abroad at conferences including, the Princeton University Economics Department, the annual meeting of the Society for Economic Dynamics, the National Bureau of Economic Research Summer Institute, and the meetings of the Midwest Macroeconomics Association, among several other venues.
Importantly, the research is informing much-needed conversations over whether economists and policymakers should favor complex quantitative models or simple but elegant methods to identify the best way to tax the wealthiest in our society.