Friday, January 5, 2007
Posted by Jeff Lipshaw
June Carbone (left) and Bill Black (right), both of UMKC, and I were trading e-mails two days ago, and to make a long story short, on pretty short notice, I filled in (to substitute for a cancellation) on a panel in the Section on Socio-Economics workshop here at the AALS meeting on Wednesday afternoon, moderated by June, and on which Bill was presenting.
The general theme of the panel was "norm-creation." Bill's talk centered on a 2005 presentation by Michael Jensen, of SSRN and all sorts of other fame, about what Jensen now sees as "low-integrity relations" between firms and analysts on the subject of earnings smoothing. I have written and posted on the relationship between law and business ethics, so I also was interested in the Jensen piece.
Since my quickly prepared presentation consists presently of what I scrawled at lunch on some LexisNexis note paper, I thought this would be a good place to preserve this somewhat impromptu "symposium" offering.
Jensen's observations are thought-provoking, particularly if you have been on the inside of a corporation making decisions about how you report your earnings. Bill is a criminologist, and his piece was about what the criminologists call "neutralization" and what I would call "co-optation, in this instance into the creation of norms under which the manipulation of accounting numbers was acceptable.
My limited goal was to take a deeper dive into how we decide something is manipulation worthy of the name "lie" or "fraud." Regular readers of this blog are, I believe, familiar with my long history as a GC at the corporate and divisional level in companies that aspired (because of the career history of the managers) to something resembling GE management style. At AlliedSignal under Larry Bossidy, the mantra every year was "Make the Numbers," a shorthand (I came to believe poorly worded) for the values of "fulfill your commitments, do what you promise, and do that for customers, employees, shareholders." So if that was one end of the continuum driving the development of internal norms of behavior, the unacceptable other end of the continuum would have been "Make Up the Numbers."
There is an epistemological element to all of this, I'm sorry to say. Accounting, in many respects, is about buckets of time, quarters and years, most of which are arbitrary (or at least as arbitrary as the fact of the Gregorian calendar and its divisions). A goal of accounting (and I'm pretty sure I could pull up a basic accounting text on this) is to match revenues and costs properly in each bucket. Smoothing is the phenomenon by which companies deliberately manipulate the revenues and costs in the various buckets so as to conform to earlier predictions, either from management or analysts, about the result in the time periods represented by the buckets.
More on this below the fold.
In the pre-Enron period, there is no question that analysts demanded, and companies delivered, if they could, no surprises from what the companies issued as their own earnings expectations for future periods.* That was the point of smoothing. Bill had an interesting thesis about those days: if the company's stock was punished because it missed an estimate by a penny (out of saying several dollars per share of earnings), it was because the market perceived that the company had exhausted every possible "fraud and manipulation" and still couldn't get to the number. I disagree, on further reflection, with that causal explanation. I think the market expected you could always manipulate another penny, so that if you missed by a penny, it was a deliberate bearish signal by management on the future prospects.
But the present question is what it means to put the terms "lie" or "fraud" as descriptors on that manipulation. I want to put aside the straw man of straight cooking the books in the manner by which I now confess I did my freshman chemistry lab reports: if you don't like the number, erase it and put in a new one. Booking sales you never made is out and out fraud. Simply changing entries you don't like is out and out fraud. Writing an earlier date on an option agreement and pretending it was signed then is a lie. Those cases, it seems to me, are too easy to be interesting.
Here's the epistemology. If a lie is a sentence uttered deliberately not to reflect reality, and with the intention of deceiving in the process, what does it means to lie about your accounting when you are talking about manipulation that is not out and out falsification of a piece of data? A financial statement is itself a model seeking to represent another reality - the state of a business. That reality is so complex that we need to reflect it in several ways, with a snap shot view at a moment in time (the balance sheet), and in flow over periods (income and cash flow statements). Lots of aspects of accounting conventions are precisely that: conventions that are proxies and do not themselves reflect reality. If you use a depreciation method, you are not really reflect the extent to which the asset is used up; you are reflecting a model of that use. And sometimes, the accounting or tax rules sanction what seems like a lie: accelerated depreciation. So what are truth statements in the context of accounting?
Moreover, the accounting conventions are subject to interpretation and judgment. For example, is the cost fairly attributable to one bucket or more than one bucket (i.e., do you expense or capitalize the cost?)
So there is something of a gray area on that continuum, between "Make the Numbers" and "Make Up the Numbers," in which we have to struggle with questions of the very essence of truth.
Here are some very cursory hypotheses:
1. Certainly pre-Enron, the rules of the manager-analyst game rewarded present period "making the numbers" over long-term value, or at least that's how companies perceived it. Woe betide the R&D department in the fourth quarter of a company having a bad year. Even without manipulation of the accounting, as Jensen observes, there was a double-think rationalization (in my view) of perfectly legal, but economically nonsensical trading of long-term value for short-term gain. See Larry Ribstein for why this supports the thesis that firms are turning to the private capital markets.
2. There is more transparency now than there used to be. That is partly related to attitudinal shifts, and partly due to the Sarbanes-Oxley rules (Regulation G) requiring there to be a reconciliation in publicly released financial statements between GAAP numbers and "as adjusted for continuing operations" numbers.
3. The integrity issues related to smoothing were not restricted to the relationship between the firm, on one hand, and securities markets, on the other. In large and complex organizations there is gaming up and down the business: business unit controllers game the division, and divisional controllers game the corporation. Jensen has another paper (only downloaded about 7,500 times) entitled Paying People to Lie: The Truth About the Budgeting System. Here is the abstract:
This paper analyzes the counterproductive effects associated with using budgets or targets in an organization's performance measurement and compensation systems. Paying people on the basis of how their performance relates to a budget or target causes people to game the system and in doing so to destroy value in two main ways: 1. both superiors and subordinates lie in the formulation of budgets and therefore gut the budgeting process of the critical unbiased information that is required to coordinate the activities of disparate parts of an organization, and 2. they game the realization of the budgets or targets and in doing so destroy value for their organizations. Although most managers and analysts understand that budget gaming is widespread, few understand the huge costs it imposes on organizations and how to lower them.
My purpose in this paper is to explain exactly how this happens and how managers and firms can stop this counterproductive cycle. The key lies not in destroying the budgeting systems, but in changing the way organizations pay people. In particular to stop this highly counterproductive behavior we must stop using budgets or targets in the compensation formulas and promotion systems for employees and managers. This means taking all kinks, discontinuities and non-linearities out of the pay-for-performance profile of each employee and manager. Such purely linear compensation formulas provide no incentives to lie, or to withhold and distort information, or to game the system.
While the evidence on the costs of these systems is not extensive, I believe that solving the problems could easily result in large productivity and value increases - sometimes as much as 50 to 100% improvements in productivity. I believe the less intensive reliance on such budget/target systems is an important cause of the increased productivity of entrepreneurial and LBO firms. Moreover, eliminating budget/target-induced gaming from the management system will eliminate one of the major forces leading to the general loss of integrity in organizations. People are taught to lie in these pervasive budgeting systems because if they tell the truth they often get punished and if they lie they get rewarded. Once taught to lie in this system people generally cannot help but extend that behavior to all sorts of other relationships in the organization.
For what it's worth, I watched this happen. I am not convinced that people respond so directly to compensation that changing the pay system would solve the problem, but I have no doubt that Jensen correctly identifies a corrupting influence from a "top-down" imposed budgeting system. I have this intuition that the gaming is more complex than merely economic. Once you set the rules of the game for success-oriented people, success-oriented people want to win. Or they want to get an A and not a C. Period.
* * *
This is about norms and integrity. My guess is the number of people who walk into these situations with the preconceived notion they are knowingly going to scheme is fairly small. That is, the set of true evil actors is relatively small. The set of banal evil, of cooptation, or neutralization, as Bill Black put, seems to me is not only bigger, but more interesting and important. And I've written about the dangers of the instrumental reasoning process by which we can delude or deceive ourselves into justifying the abuse, all of which can be exacerbated by a lawyer's professional gloss (if not imprimatur) on the justification.
I don't think the solutions are algorithmic. I am suspicious of instrumental reason (or instrumental reason masquerading as pure practical reason). I am aware of the mushiness of relying on intuition. So it's a mystery to me still how we resolve the intersection of legal rationalization with a moral and ethical sense.
*(The practice of issuing "guidance," as it is called, has substantially curtailed since then, I think. I saw some data just a few days ago that securities class action filings are down - that would be consistent with less earnings guidance - the core of a archetypal suit consists of company guidance and then a subsequent event that proves the guidance incorrect.)