Governance through Shame and Aspiration: Index Creation and Corporate Behavior in Japan

Charles C.Y. Wang is the Glenn and Mary Jane Creamer Associate Professor of Business Administration at Harvard Business School; Akash Chattopadhyay is an Assistant Professor of Accounting at University of Toronto Mississauga and the Rotman School of Management; and Matthew Shaffer is a doctoral student at Harvard Business School. This post is based on a recent paper by Professor Chattopadhyay, Mr. Shaffer, and Professor Wang.

There is growing interest in using stock indexes to shape corporate behavior and the standards of corporate governance. Over the past weeks, two of the largest index providers—S&P Dow Jones and FTSE Russell—announced their decisions to exclude certain firms with multiple share-class structures from their indexes. Despite these significant moves, empirical research has not established whether and how stock indexes can be effective in shaping standards of corporate behavior.

In a new working paper recently posted on SSRN, Governance through Shame and Aspiration: Index Creation and Corporate Behavior in Japan, we examine the governance role of stock indexes by exploiting the unique features of Japan’s JPX-Nikkei 400 index (JPX400). Launched in 2014, the JPX400 consists of the 400 best performing firms in terms of profitability among Japan’s largest and most liquid firms. Our analysis shows that the index had profound effects on Japanese firms and the overall stock market, and that these effects were predominantly driven by managers’ prestige concerns—the aspiration to acquire prestige or the desire to avoid shame—rather than the financial benefits of index inclusion.

The Japan Exchange Group (JPX) and Nikkei, Inc. designed the JPX400 to showcase firms with high profitability and capital efficiency. Every summer, the JPX ranked Japanese firms on a composite measure based on their ROE, operating income, and market capitalization, and selected the top 400 firms for membership in the index over the subsequent year. The index echoed the concerns and goals of Japanese policymakers, who, under the Abe administration, saw Japanese corporations’ lagging capital efficiency (as measured by ROE and ROA) and lack of engagement with shareholders as important contributors to the country’s ongoing macroeconomic malaise. With the endorsement of the GPIF (which committed to track the index) and the LDP, the index quickly acquired prestige status, and adopted the nickname “the shame index,” a reference to the experience of firms that failed to make the cut each year. The JPX400’s yearly inclusions and exclusions generated substantial media attention, with many firms that had missed the index publicly committing to change their corporate policies in the hopes of making it into the index in future years.

To study the effect of the index on Japanese corporate performance, our identification strategy exploits the variation in the “intensity” of index-inclusion incentives between firms. Because firms closer to the inclusion threshold are most likely to see their inclusion status change as a function of their performance, all else equal, firms closer to the threshold of inclusion should experience the most intense incentives to improve their performance to gain membership in (or avoid expulsion from) the index.

While the JPX does not publish its rankings of firms, it uses a transparent algorithm for selecting members, which we are able to replicate. We show that our rankings can predict JPX400 membership with a high degree of accuracy, and use them to measure the variation in firms’ ex-ante index-inclusion incentive intensity. We then compare those firms nearest to the threshold of inclusion/exclusion to those further from the threshold. More specifically, firms within a 100-firm bandwidth of the rank-400 cutoff for index inclusion have a substantial probability of seeing their index membership status change in the next year, while firms outside this bandwidth have a low probability of changing their membership status. Thus, the incentives to be included in the index primarily, and most strongly, affected firms ranked 301-500. We use these firms as the treatment group, and use firms ranked 501-800, which are fairly similar in their background characteristics but have insignificant probability of JPX400 inclusion, as our controls. Moreover, due to the transparent nature of the algorithm, we can assign synthetic or placebo ranks to firms in the years before the JPX400 index actually existed. This means that we can estimate the effect of the index using a difference-in-differences (DID) regression, which compares the outcomes for the treatment vs. control group in the post-period relative to the difference in their outcomes in the period before the index existed.

Our main outcome of interest is ROE because it was the distinctive focus of the index and the most controllable determinant of firms’ ranks. In our main regressions, we find that “treated” firms experienced an increase in ROE of 1.9 percentage points, a 35% increase relative to their pre-period means. This effect is robust to controls, fixed effects, alternate specifications, and placebo tests. Aggregating across the 200 treated firms, the JPX400’s total effect is substantial: it accounts for 23% of the average increase in aggregate annual earnings over our sample period and a 3% increase in total market capitalization.

Using variations on this research design, we then proceed to examine 1) why firms were motivated to work for inclusion in the index; 2) what channels they exploited to achieve such strong improvements in ROE; and, 3) the effects on other outcomes of interest to Japanese policymakers, the index creators, and other stakeholders.

First, we test whether prestige or capital-market benefits drive our main findings. We separate these two explanations by comparing the response of treatment firms that were members of the Nikkei225 to those who were not. Nikkei225 members, by definition, already have access to the financial benefits of inclusion in a major, liquid, and highly-tracked index. Thus, these firms would have smaller prospective capital-market benefits (e.g., lower cost of capital) from inclusion in the JPX400; but, as publicly prominent, well-known firms, they would experience stronger degree of embarrassment (or shame) from exclusion. We find that the treatment response is substantially larger for Nikkei225 firms. A triple-difference design also produce similar results, suggesting that prestige-based concerns—the aspiration to attain prestige status or the (potential) embarrassment/shame due to loss of prestige—were the main drivers of firms’ JPX400-inclusion incentives.

Second, we decompose ROE into profit margin, asset turnover, and leverage, in order to examine the channels our treated firms used to improve ROE. We find that the overall effect in ROE was driven by operational improvements, in particular through ROA and profit margins, rather than financial engineering. However, our subsample analyses revealed that firms sought improvements where they had more slack. For example, firms with relatively high levels of cash substantially increased shareholder payouts, and relatively inefficient firms increased asset turnover substantially.

Third, we evaluate the effects of the index on other outcomes that would likely be of interest to policymakers and index-providers interested in assessing the overall effects of the index. We find that the JPX400 increased treated firms’ valuations (as measured by Book to Market ratios), but had no statistically detectable effect on accruals (a proxy for earnings management), capital expenditures, R&D expenses, or total employment.

Collectively, our results provide robust evidence that a stock index can serve as a source of prestige that can be actively employed to motivate corporations to change their behavior. We thus provide a fresh perspective to the corporate governance literature, which has traditionally focused on using formal contracts and pecuniary incentives to motivate managers; and to the index literature, which has largely focused on the effects of index inclusion on firms’ subsequent behaviors, rather than firms’ efforts to gain inclusion into (or avoid exclusion from) indexes. Our findings could be especially important to policymakers and boards in the future, as they become increasingly constrained in using executive compensation to motivate and govern managers, and look for novel and powerful channels of corporate governance.

Our finding that prestigious indexes can be powerful tools of corporate governance is especially timely and germane to S&P Dow Jones and FTSE Russell’s recent decision to exclude certain companies with multiple-voting class structures from their indexes. Specifically, we highlight an important (and perhaps surprising) reason why indexes can motivate and influence changes in corporate behavior: managers’ concerns for prestige.

The complete paper is available for download here.

Both comments and trackbacks are currently closed.