Better ways to use synthetic controls to model state and local tax policy

The synthetic control method (SCM) is a relatively new and valuable tool that makes it possible to measure state or local policy against a theoretical baseline, when an actual basis for comparison does not exist. Because technology is constantly evolving, we wrote “An update on the synthetic control method as a tool for understanding state policyIt includes some changes that improve the SCM, as well as some graphic intuition behind the method and a new way of looking at how to select the ideal pool of possible controls.

SCM was popularized by MIT economist Alberto Abadi and his colleagues in 2010, when they used Estimate how a tobacco tax increase in California affected cigarette purchases, Some of its appeal comes from its ability to combine the clarity and transparency of case studies with the rigors of statistical analysis. The model has been widely used and many improvements have been added over the past decade.

In 2017, Rob and former TPC staff member Sarah Gault wrote “Synthetic control method as a tool for understanding state policyOur new report updates that 2017 paper.

Analysts can use the SCM to evaluate policies that affect only one “unit” or jurisdiction, such as a country, state, or city, where there is no clear control against which its experience can be compared. . They create a synthetic control by adding and weighting a small number of jurisdictions that are similar to the area they are studying.

If synthetic control is sufficiently similar, researchers can understand the impact of a policy change by plotting the difference between the actual jurisdiction and the results of synthetic control. For example, in 2010 Abadi and his colleagues estimated the effect of a 25 percent per pack increase in tobacco taxes in California, comparing the sales of cigarettes in that state with their synthetic sales in California.

Before the 1988 tax increase, per capita cigarette sales in synthetic California closely matched sales in real California. But after the tax hike, real sales fell faster than synthetic California. This difference reflects the estimated impact of the tax increase.

Our new report revises the TPC’s 2017 guide to take into account recent improvements to the SCM. For example, it now shows how to reduce interpolation bias, which can occur when the jurisdictions that make up the synthetic controls are not the same as the jurisdictions with policy changes. Perhaps unsurprisingly, this can happen even when synthetic controls are the same as policy change jurisdictions.

Although analysts often prefer large datasets, using a very large pool of potential comparison jurisdictions with SCM may also bias estimates of policy effectiveness. There are complicated ways to fix this problem, but we show a simple graphical method. We also provide links to computer code incorporating some model improvements.

Given the current popularity of the SCM, it is increasingly likely to be used by analysts. Our updated guide provides an easy-to-understand explanation of improvements to the SCM that could make the model more useful by researchers and policy analysts.