Do I need to include control variables when using a matched sample

[Felipe Calvano]

I would like to know if I need to include the variables that I matched the companies as control variables. In my data, I created a matched data set based on firm size and industry.
Also, I have seem some studies that did not include control variables at all after they created the matched sample. Is that correct?

Thank you

[Phil Bromiley]

The answer depends on how good your match is. If there are variables on which your match is not very good, then control variables make sense. I have had an occasion where I matched on size, but still found size influenced the outcome (even with fixed effects for each matched pair). In general, you shouldn't lose much by including them as controls.

[David Benson]

Hi Felipe, I suspect that just matching on size and industry may not give you a close enough match, so you will probably want to include other control variables. But I've included several finance / accounting / entrep papers that used matched samples (but they also often used fixed-effects regression, regression discontinuity, or 2SLS to triangulate their results). They cover a variety of settings (bankruptcy, acquisitions, startups receiving venture capital, banks making loans, etc). Some included controls in their matched sample, some didn't.


Finance & Accounting:

• Financial Reporting Quality of U.S. Private and Public Firms, Link
• Financial Reporting Frequency, Information Asymmetry, and the Cost of Equity (it's been published, but ungated WP version here
• Do Acquisitions Relieve Target Firms’ Financial Constraints? Link
• Capital ratios and bank lending: A matched bank approach Link
• Predicting corporate financial distress: Reflections on choice-based sample bias Link

The last paper argues that matched samples can still be biased:

ABSTRACT: Financial distress precedes bankruptcy. Most financial distress models actually rely on bankruptcy data, which is easier to obtain. We obtained a dataset of financially distressed but not yet bankrupt companies supplying a major auto manufacturer. An early warning model successfully discriminated between these distressed companies and a second group of similar but healthy companies. Previous researchers argue the matched-sample design, on which some earlier models were built, causes bias. To test for bias, the dataset was partitioned into smaller samples that approach equal groupings. We statistically confirm the presence of a bias and describe its impact on estimated classification rates.

Entrepreneurship

• The Consequences of Entrepreneurial Finance: Evidence from Angel Financings, The Review of Financial Studies, Volume 27, Issue 1, 1 January 2014, Pages 20–55, https://doi.org/10.1093/rfs/hhr098
• How Does Venture Capital Financing Improve Efficiency in Private Firms? A Look Beneath the Surface. The Review of Financial Studies, Volume 24, Issue 12, 1 December 2011, Pages 4037–4090, https://doi.org/10.1093/rfs/hhr096

[Felipe Calvano]

Thank you very much!