TPC is about to begin releasing its estimates of the revenue and distributional effects of the 2020 presidential candidates’ tax proposals, a project we’ve done every four years since 2004. Before we release our analyses, I’d like to explain how we do it.

Which plans do we model?

Given the large number of presidential candidates and our limited staff time, we cannot model every candidate’s plan. Generally, we use four criteria to choose what to analyze:

• Has the campaign provided sufficient detail in describing their tax plans?
• Can we accurately model the proposal?
• Does the proposal present an opportunity to discuss an important or novel concept?
• Does the proposal have a noticeable influence on the tax policy debate?

How do we model proposals?

Step 1: What we know (and don’t know)

We start with the tax plan descriptions provided by each candidate on their websites and in their public statements.

Step 2: Distributional and revenue analysis

We then ask each campaign for additional information to help us analyze its plan. If we don’t get sufficient detail, we make assumptions to fill in the blanks. For example, we may rely on similar pre-existing legislative proposals. We explicitly describe any assumptions we make and may also discuss the implications of alternative assumptions. We may revise our analysis if campaigns provide additional details at a later date.

We then use our microsimulation tax model to analyze the revenue and distribution effects of each plan.  The model, which we have been using and refining for more than a decade, uses data from the IRS and other sources.

It shows how proposed tax changes will affect households of different incomes and produces a traditional/conventional revenue estimate. That analysis reflects how taxpayers change their economic behavior in response to tax changes (such as changing saving patterns or reducing purchases of taxed goods), but does not account for how a tax plan changes the overall economy.

Generally, we will follow-up our traditional analysis with a macroeconomic analysis that does include the effects of tax changes on the overall economy. However, because we first must know the distributional and revenue effects, the macroeconomic analysis will come later.

Once we’ve modeled a candidate’s plan, we write a report that is reviewed by our senior staff as well as a politically diverse group of outside experts. Once the analysis is complete, we make it and supporting data available on our website.

Step 3. Additional analysis

After publishing our initial revenue and distributional analysis, we may supplement that work with some of the following components:

• The impact of the proposal’s effective marginal individual income tax rates on capital and labor income
• The impact of marginal effective tax rates on new investment
• Effects on the federal budget deficit and debt over time
• A qualitative analysis of how a candidate’s proposal compares to other policies designed to achieve the same objectives and whether it achieves principles of sound tax policy

The work is challenging, but our goal is simple: to provide voters with an objective, non-partisan, data-driven analysis of the presidential candidates’ tax plans. To read more about how we do it, click here.