Journal of Applied & Computational Mathematics

Forecasting the US Capital Gains Tax Rate Using Markov Chains

Abstract

Brian C Trowbridge

Using Markov-chains (formally a Markov process or Markov system) we will show how to estimate the probabilities of a capital gains tax rate changing from its current rate to another rate within a certain number of years. It is important to recognize that capital gains rates in the United States have their origins in every elected administration of the government that had the power to influence tax policy. The outcome of government elections is dependent on chance factors such as periods of war and peace, economic recession and expansion, current fiscal and monetary policy, changes in population demographics, just to name a few. The delicate balance between the government and the dynamic society that shapes tax policy is difficult to predict using deterministic modeling. Therefore, we look to stochastic modeling using the Markov-Chain processes to see if it could be of use in creating a useful model for capital gains tax rates in the United States.