Do Investors Understand ‘Operational Engineering’ before Management Buyouts?

In our paper, Do Investors Understand ‘Operational Engineering’ before Management Buyouts?, which was recently made publicly available on SSRN, we use a sample of management buyouts (MBOs) from 1985-2005 and a matched subsample of post-MBO firms to examine three questions. First, we examine whether firms undertake different types of activities to lower earnings before MBOs. Second, to see whether outside investors and the market understand such ‘operational engineering’ activities, we study the impact of these activities on target firms’ stock returns and MBO deal characteristics including deal premium and likelihood of deal completion. Third, we examine the relation between pre-MBO earnings-reducing activities and the post-MBO operating performance.

With the Great Recession of 2007-2009 exposing deficiencies of the world’s most advanced financial markets, leveraged buyouts (LBOs) have ‘reemerged’ as a solution to the many challenges facing corporate sectors. Unlike publicly listed firms, LBO firms are characterized by concentrated ownership, active monitoring and high leverage. A growing strand of literature shows that LBO firms can create value through ‘financial, operational and governance engineering’ (Kaplan and Stromberg, 2009). In fact, Jensen (1989) argues that LBOs should replace publicly held corporations as the dominant corporate organizational form.

A key question studied by the prior literature is whether LBOs improve firms’ operating performance, as measured by return on assets (ROA), among others, which are based on reported earnings. In particular, a number of papers find substantial improvement in ROA following buyouts in which the target managers and other insiders participate and the authors attribute the improvement to the strong incentives for value-creation created by MBOs. However, part of the improvement in ROA may be explained by how earnings are reported both before and after the MBOs, as it has been well established that this process is subject to managerial discretion. If managers of a publicly listed firm plan to take the firm private, they have an incentive to undertake activities to temporarily deflate earnings before launching an MBO. With lower earnings before the MBO, the firm can display greater improvement in operating performance after the MBO once the earnings-reducing activities are halted. Moreover, lower pre-MBO earnings can lead to lower stock prices, enabling the buyout team to acquire the target ‘on the cheap’ if outside investors do not fully understand the nature of these activities.

In our investigation of possible earnings-reducing activities before MBOs, we rely on the extensive literature that documents different forms of ‘earnings management’ to influence the market’s perception of the operating performance of firms. The practice of earnings management is also corroborated through anonymous surveys, in which managers do admit that they engage in various forms of earnings management (e.g., Graham, Harvey and Rajgopal, 2005). We investigate earnings management based on accruals and ‘real’ activities. First, the accruals component of earnings typically includes estimates of future cash flows and deferrals of past cash flows, thus reflecting managerial judgment and can potentially be used to deflate earnings. As an example, aggressive provisions for bad debt expenses can lower a firm’s current period earnings; when the firm receives debt payments or recovers losses from bad debt at a future date, the estimations of subsequent expenses must be lowered, leading to an increase in future earnings.

The second type of earnings management does not involve any accounting choices, but entails managerial decisions in the timing or structuring of operations, investments, and/or financing activities that have cash flow consequences. Since this type of activities can be disguised as an integral part of business decisions, so that even dramatic changes can be justified as rational responses to a changing environment, they are generally more difficult to detect. For MBO targets, activities such as increasing selling, general, and administration (SG&A) expenses in the form of R&D and advertising costs, as well as losses in asset sales, can lower current period earnings. However, a return to reporting ‘normal’ levels of expenses and gains/losses in asset sales, along with possible payoffs from R&D expenditure, can help boost operating performance after the MBO.

In our first set of tests, we find that during the year prior to the MBO deal announcement date, target firms exhibit abnormally high discretionary SG&A expenses, abnormally low discretionary accruals, and realize losses from asset sales. In comparison, the average levels of all three variables are insignificant from zero for these firms two years before the MBO announcement. For the subsample of MBOs for which we obtain post-MBO data, we find no evidence of any earnings management activities after the completion of MBOs. In addition, the average levels of all the earnings management measures are insignificant from zero for the sample of matching firms – non-LBO firms matched by industry, size, and book-to-market in all three years before MBO announcements or after the MBOs. These results are consistent with the hypothesis that managers adopt earnings-decreasing activities to prepare specifically for the upcoming MBO transactions.

In the second set of tests, we begin with studying the effects of these activities on pre-MBO stock prices. We find that higher levels of discretionary (SG&A) expenses and losses from asset sales are associated with lower one-year cumulative abnormal stock returns before the MBO announcement date. These results are robust after controlling for a range of firm characteristics measuring the firm’s investment opportunities and financial conditions as well as other factors that may affect stock returns. However, no such relation exists between discretionary accruals and pre- MBO abnormal stock returns. We also do not find such a relation between the abnormal stock returns over the same period and the measures of any of the activities for the matching firms.

In cross-sectional analyses, we also obtain a number of interesting results on the relation between the level of earnings-reducing activities and pre-MBO abnormal stock returns. First, this relation is mainly driven by MBO firms whose insider ownership stakes (before the MBO) are greater than the sample median. This result supports the hypothesis that insiders who can gain a great deal from the MBO transaction have a strong incentive to employ opportunistic, earnings- reducing activities. Second, the relation is more pronounced among MBO firms with greater information asymmetry, proxied by lower analyst coverage and asset tangibility, higher levels of analysts’ earnings forecast errors and dispersions in forecasts, and greater stock return volatilities. This relation is also stronger for MBO firms with less institutional holdings. Since our measures of earnings management are derived as deviations from ‘normal’ levels of accruals, SG&A expenses and gains and losses from asset sales, recent reviews have emphasized the importance to validate these measures in situations where managers have a clear and amplified incentive to manage earnings (e.g., Dechow and Skinner, 2000; Healy and Wahlen, 1999; Schipper, 1989). In this regard, our cross-sectional results provide further evidence that lower pre-MBO stock prices are likely a result of opportunistic managerial behavior, because such behavior is more likely to be detected and/or has a smaller impact on stock prices in firms with more institutional monitoring and less information asymmetry.

We next examine whether the earnings-decreasing activities before MBO announcements affect several MBO deal characteristics. These include deal completion likelihood, announcement period returns (cumulative abnormal stock returns from announcement dates to effective dates), and deal premiums (offer price over target price four weeks prior to deal announcement dates). If outside investors begin to understand management’s actions around the time of deal announcement and react accordingly – for example, adjusting the price or threaten to reject the deals, then the effects of earnings-reducing activities on pre-MBO stock prices can be partially or fully offset. However, we find that none of the three measures affect any of the three MBO deal characteristics.

Finally, for the subsample of firms matched with post-MBO data, we confirm results from earlier research in that they also exhibit significant improvement in operating performance, as measured by ROA. More importantly, abnormal discretionary expenses (e.g., advertising expenses and R&D) and losses from asset sales, which we found to be associated with lower stock returns before MBO announcements, are positively related to post-MBO performance as measured by ROA.

Taken together, our evidence suggests that MBO target firms undertake earnings-reducing activities through discretionary expenses, losses from asset sales and discretionary accruals before MBOs. Outside investors and the market do not fully understand real activities based earnings management, especially for firms operating in environments with severe information asymmetry, little institutional monitoring and greater insider holdings. This, in turn, allows managers and other insiders to acquire the targets “on the cheap,” show higher post-MBO operating performance, and earn higher returns from the MBO transactions. Our findings confirm and extend survey-based evidence in Graham, Harvey, Rajgopal (2005) that managers use real activities based earnings management and that they prefer these activities to accruals based management. Our paper contributes to the literature on MBOs and LBOs in general by providing an additional explanation to the value creation of buyouts. Future work on the performance of MBOs should take into account of the effects of ‘operational engineering’ activities before deal announcement.

The full paper is available for download here.