Stata Code for zero-leverage firms and reference set of proxy firms

Ilya A. Strebulaev and Baozhong Yang (2013)  report that almost 22% of public nonfinancial firms in the US have almost zero leverage.  They report that these firms have unique characteristics such as  Dividend-paying zero-leverage firms pay substantially higher dividends, are more profitable, pay higher
taxes, issue less equity, and have higher cash balances than control firms chosen by industry and size.

Where the authors’ propositions are simple, their research design is not. In their empirical setting, they carefully select zero-leverage (ZL),  almost zero leverage (AZL) firms and a reference set of proxy firm-years. In the benchmark construction procedure, they select proxy firms by calendar year, industry, size, and dividend-paying status. Then the ZL and AZL firms are compared with their benchmarks with respect to book leverage, market leverage, log(size), market to book , cash, profitability, dividends, tangibility, tax, age, capital expenditure, asset sale, equity issuance, retained earnings, and non-debt tax shield. The comparisons are done through two-sample t-tests.


Our Stata Code

We have developed an efficient code for constructing the sample of zero-leverage, almost zero-leverage and comparable proxy firms. The code is extremely efficient in making the list of comparable proxy firms for ZL and AZL firms. Specifically, in each industry-firm year combination, our code finds four proxy firms for each focal firm and then finds a cross-sectional average of the relevant firm-specific variables for the proxy firms. As a next step, one can easily apply t-test to see whether the two sets of firms differ.



The code is available for $299, plus $50 for raw data processing (in case the data is not in Stata format and variables are not already constructed). For further details, please contact us at: