The Double Dividend of Relative Auditing – Theory and Experiments on Corporate Tax Enforcement

Recent papers have shown that theoretically tax authorities can not only reduce tax evasion but also boost output in oligopolies by conditioning the audit effort spent on a firm on all firms' tax returns in an industry. In this paper we revisit these results and extend the class of relative audit rules with this property by including discontinuous rules. Field experiments testing the theory predictions would require randomizing audit rules across many otherwise identical industries and are therefore impractical. Instead we conduct laboratory tests of the theoretical mechanisms of a variety of rules. We find that both dividends of relative auditing, i.e. less evasion and higher output, materialize in the laboratory. The behavioral mechanism generating the higher output differs somewhat from the one propagated by theory though.