08 April 2010

Exit, Voice and Repairable Lapses

Economists tend not to think much about "repairable lapses" i.e. "declines [in organizational performance] that could be corrected with the right balance of information, incentives and flexibility of response." Rajiv Sethi, discussing the work of economist Albert Hischman, explains that there are serious downsides to this attitude.

Most economists assume most of the time that market actors are rational (i.e. that they don't have lapses) and that the market will weed out firms that lapse too badly. This is unrealistic.

[T]he vision of a "relentlessly taut economy" operating at or close to its productive potential is inapplicable to technologically modern societies capable of producing a substantial surplus relative to the needs of subsistence. The very existence of the surplus implies that considerable slack in the level of efficiency can be tolerated without disastrous consequences. As a result, firms and other organizations are "permanently and randomly subject to decline and decay, that is, to a gradual loss of rationality, efficiency, and surplus-producing energy no matter how well the institutional framework within which they function is designed."


So what are the downsides of not having alternative to economic survival of the fittest through firm failure to deal with repairable lapses?

First, since most firms can continue to operate despite medicore performance, their mediocrity does more to make our economy suboptimal than firms whose performance can be summed up as "total fail."

Second, when a firm fails due to a repairable lapse, there is a risk that it will send the wrong signal, suggesting that a business model or economic niche is flawed when the real problem is a management screw up that wasn't corrected. This misinformation can lead to long term economic harm.

Consider the Hindenburg. A deadly 1937 airship accident ended the airship industry, probably due to the use of a flamable paint used on the fabric skin and the use of hydrogen rather than helium to fill the airship. Had a knowledgable person identified the problems, or the way the coverage happened had been less compelling, the design flaws could have been solved and airships could have slowly declined as fixed wing airplane technology improved and airships would probably still have important niche markets in areas where building roads is expensive. Instead, it has taken more than sixty years for anyone to seriously reconsider commercial airships in the niches where they make sense.

Third, we live in an age of big businesses in which the harm of letting a firm fail can be great. The "too big to fail" concept is real, but in real life it plays out on a continuum, rather than than all of nothing. The bigger the average business is, the more likely it is that a sink or swim attitude towards firms will harm the economy.

Often when a big economically important firm, such as the former big accounting firm Arthur Andersen, simply fails due to what was probably an isolated mistake (one partner of thousands in a branch office shredded documents, producing a criminal conviction that was later reversed on appeal), the market won't necessarily produce a prompt replacement. In the meantime, everyone who relied on the failed firm must incur often substantial costs to seek the same services elsewhere, and the reduced competition in the market allows the remaining firms in the market to increase their prices without offering additional value.

There are different ways that economic actors can respond to repairable lapses. They are called in economics shorthand exit and voice. For example, from an investor's perspective, "exit" is epitomized by the "Wall Street Rule" of selling shares in companies you think are poorly managed, while "voice" is epitomized by trying to get a seat on the board of directors so that management can be replaced. More generally, those who think that management is screwing up can offer constructive criticism.

One problem when exit is too easy is that it can undermine an otherwise repairable lapse if it is exercised to quickly to allow a solution to be formulated. The classic example is a run on a bank. A few quiet complaints and slow reduction of deposits due to some concern can be solved by raising more capital, or identify a problem line of loans that threatens to undermine the firm as a whole. But, if everyone withdraws their funds at once, the bank dies before it can correct the error of its ways.

A major justification for "bailouts" public or private, and for bankruptcy reorganizations is that they can prevent rapid exit by dissatisfied economic actors from turning a repairable lapse into an irrepairable one.

Another key observation in this framework is that those most like to make effective use of "voice" were exit not an option are also often those most likely to exit if there is a not unduly costly alternative.

Public schools might deliver worse learning outcomes if private or parochial options are available to the most quality conscious parents. A small decline in neighborhood quality could turn into a precipitous collapse if those most affected by it simply move elsewhere. And the ease with which common stock can be sold implies that the most vigilant shareholders will liquidate their holdings rather than attempt to improve the performance of management.


Indeed, near monopoly holders often want to have some competition "in order to shed their most vociferous customers" who might otherwise push to wipe out the slack that management gets to enjoy.

Increasing the cost of exit and developing loyalty to an organization can strengthen the quality of "voice" signals that a firm receives, allowing it to solve its repairable lapses before they get out of hand and benefitting everyone who deals with the organization, and the economy as a whole.

One reason that rural schools, for example, may do a good job with the resources that they have available to them, is that there are no alternatives, so even very quality conscious parents need to make them work. Indeed, rural civil society in general are empirically particularly sucessful at using democratically organized consumer and producer cooperatives to meet their needs. In part, this is because economies of scale mean that it is only possible to have one firm of a particular type in an area. This makes the cost of exit very high for any essential service provider, and makes the value of voice in solving repairable lapses great.

The concepts have broad applicability in conceptualizing public policy issues. For example, does reducing the cost of exit from a marriage though "no fault" divorce reduce the incentive to use means short of dissolving the marriage to address repairable lapses?

Fault based divorce sought to address the problem by distinguishing between irrepairable lapses (fault grounds) and repairable ones. The courts turned out to do this job rather poorly, and for that reason, no fault divorce has swept the nation with bipartisan support. But, this doesn't mean that the conceptual framework that guided the middle ground of fault based divorce that followed legislative divorce and was followed by no fault divorce couldn't helpfully inform efforts to reduce the harm that divorces in marriages that could have been saved with the right kind of intervention cause.

No comments: