Florian Hoffmann, Vladimir Vladimirov
Based on: “Worker Runs,” forthcoming in the Journal of Finance. DOI: https://dx.doi.org/10.2139/ssrn.4150240
In June 2021, the financial press reported extensively on Credit Suisse’s fight to stem the exodus of senior and junior employees across multiple divisions. The trigger for these departures was the bank’s exposure to the spectacular collapses of Archegos and Greensill Capital, which dented its profits. Senior bankers not involved in these affairs, whose divisions would otherwise have led the bank to a record quarterly profit, were reportedly furious and left as a result. Other bankers then started to leave in droves, as “nobody wanted to be the last man standing.” Pundits worried that the snowball effect of such departures could seriously erode dealmaking and the bank’s market position.
The turnover contagion faced by Credit Suisse illustrates a broader phenomenon widely discussed by academics and industry practitioners. A recent study by Visier of over 10 million workers, which quantifies the problem of contagious turnover, documents that an initial resignation in a team is associated with a 9% higher probability of further resignations within the same team. A better understanding of such contagion effects is crucial, given that over the last 20 years, 23% of workers have voluntarily quit their jobs each year. The cost to firms is staggering. In the U.S. alone, the annual cost of replacement, training, and lost productivity is reported at over a trillion dollars. In 2014, Credit Suisse estimated that reducing turnover could save it $100 million per year. Some of these costs are inevitable, as employee turnover is part of a natural and efficient process of workers finding a better match for their skills. However, it is clearly in firms’ interest to lower the cost of inefficient contagious turnover — i.e., the turnover resulting from a coordination failure among workers in fundamentally sound firms. Existing recommendations to mitigate this problem range from improving company culture to increasing the level of pay, but concrete prescriptions are few. In this paper, we consider the structure of pay as an additional remedy.
Worker runs. Key factors affecting workers’ decision to leave a firm are current compensation and concerns about future compensation. One such concern is that the departure of hard-to-replace workers reduces firm productivity. That, in turn, reduces the value of the remaining workers’ compensation linked to firm performance, such as equity-based compensation or bonuses. As a result, the voluntary departure of key personnel can trigger remaining workers to follow suit, resulting in “worker runs.” Worker runs are particularly likely and costly in firms that rely heavily on teams of hard-to-replace workers with complementary skills. Typical examples are startups whose success depends crucially on retaining well-functioning teams of skilled employees in R&D, management, sales, and marketing. Other examples include PE partnerships and consulting, advisory, investment banking, and law firms. Large firms such as Credit Suisse are also affected, with contagion being two to three times higher when they are organized around smaller teams.
1. Deferred fixed pay. A firm offering fixed pay to its workers is already going a long way toward preventing contagious turnover; deferring fixed pay is even more effective in improving retention.[1] The key advantage of fixed compensation is that its value, unlike that of bonuses or equity-based compensation, does not depend on the retention of other workers (as long as the firm can honor its promises). This ensures that employees worry less about their promised future compensation losing value when their coworkers are leaving. Thus, the remaining employees are less likely to run.
2. Dilutable compensation. Offering high enough fixed compensation to guarantee retention may conflict with other compensation objectives, such as offering performance bonuses to incentivize employees. Furthermore, fixed pay may not always be feasible or cost-efficient for firms. For example, cash-constrained firms such as startups may be unable to promise high fixed salaries. Moreover, firms may want to offer equity compensation to better tie employees’ compensation to industry dynamics and, thus, workers’ outside options.
When firms rely more on compensation tied to firm success, such as equity-based pay or performance bonuses, they can lower the cost of collective retention and mitigate the risk of worker runs by making compensation “dilutable.” In other words, firms can implicitly or explicitly promise remaining workers higher pay when other workers are leaving.
A simple example illustrates the concept. Suppose a firm with 100 outstanding shares sets aside 100 additional shares for compensation purposes. For simplicity, suppose that these shares are promised to two workers, each of whom receives 50 (time-vesting) shares if she stays with the firm for a specified period. If both workers stay, each worker will own 50/200=25% of the firm’s equity. However, if one worker leaves and forfeits her equity compensation of 50 shares, the remaining worker’s equity stake will increase to 50/150=33%. The two advantages of such “dilutable compensation” are now immediate. First, even though the firm’s productivity and equity value may drop if one worker leaves, the higher percentage of equity ownership counteracts the other worker’s incentive to leave. Second, if the firm manages to retain both workers, the higher retention level dilutes each worker’s percentage equity ownership, allowing the firm to retain the workers at a lower cost. As this example illustrates, deferred equity-based compensation improves both individual and collective retention by implicitly tying compensation to the firm’s overall retention level. Equity buy-back agreements introduce a similar dilution feature to compensation. Importantly, dilution is by far not restricted to equity-based pay, and it can be easily decoupled from firm size. For example, firms may offer profit-sharing bonus pools at the division or team level. Furthermore, offering retention bonuses can also effectively make compensation dilutable.
3. Paying employees differently. Another way to reduce the risk of worker runs is to introduce pay heterogeneity, which means that individual members of teams of workers are paid differently, not just in terms of level but also structure. This is particularly attractive if heterogeneity arises naturally because employees have different skill sets or experience. However, it is useful even if workers have similar skills. In either case, targeting a subset of workers with compensation that makes them willing to stay for sure means that all other workers do not need to worry as much about the firm’s overall retention. As a result, they are also more likely to stay and are cheaper to retain. Notably, ensuring that some workers always stay does not necessarily require paying them more if the firm appropriately designs the structure of pay.
To illustrate these points, consider a firm trying to retain two workers, Alice and Bob, each demanding $100k per year, but the firm’s overall safe cash flow is only $100k. Hence, if the firm wants to pay both workers the same, it needs to supplement its fixed wage offers with performance bonuses or equity-based pay. However, such variable compensation depends on the firm’s success and thus on its ability to retain both of its productive workers. Thus, Alice and Bob will demand a high bonus or equity stake to account for the risk that the firm’s probability of success diminishes if the other worker leaves. To avoid this, the firm can use its limited resources to guarantee Alice a payment of $100k via a fixed wage or a retention bonus. Because the firm is constrained by its available riskless cash flow, it cannot make the same fixed pay offer to Bob and, in this stylized example, needs to pay him entirely with variable compensation, the value of which depends on Alice’s retention. However, since Alice’s fixed pay is high enough to ensure that she always stays, Bob does not have to be compensated for the risk that Alice leaves. This lowers the expected compensation that Bob has to be promised over his outside option of $100k (what Alice receives) and, thus, the firm’s overall retention costs. More generally, the key advantage of paying some employees differently is that it lowers retention costs without introducing substantial differences in pay levels. It is particularly attractive if a firm wants to improve its probability of retaining certain employees. Such differences in payment structure could, in practice, be motivated by awarding workers somewhat different job titles, such as “associate” and “senior associate.”
As these examples illustrate, the structure of compensation plays a key role in mitigating “worker runs.” The tools — equity-based pay, profit-sharing bonuses, retention bonuses, and structuring compensation differently (e.g., for different job titles) — are all common in practice. Reducing the risk of contagious turnover is a matter of improving our understanding of how best to use them
[1] Though practitioners often associate deferred compensation with vesting equity-based pay, any compensation can be deferred to improve its retention properties (for example, think of annual bonuses in U.S. finance and law firms, which are often implicitly guaranteed).