Asymmetrical Information

From the desk of Jane Galt:

Outstanding: Mindles Dreck has some

Outstanding: Mindles Dreck has some suggestions for real reform of the way we regulate disclosure, not just handing the SEC more money and eliminating the expensing of stock options. And so I've decided to make a list of my own: reasons why just increasing the budget of the SEC will not magically make companies honest.

1) Regulatory Capture (follow the link -- one of the most outstanding "explanation" posts of all time) The SEC, like every other regulatory agency in history, eventually comes to serve the interests of the industry it regulates rather than the public at large.

2) The investment banks and their law firms have more money than the SEC The main enforcement guys at the SEC are lawyers. I don't know how much they make, but it's some embarassing sum lower than 40K a year. How much lawyer do you think you could buy for 40K per annum? Yup, the market works. And that means that the lawyers thinking up ways to evade the regulations are going to be, on average, smarter, more driven, and working longer hours, than the lawyers trying to enforce them.

Raise their salaries to the level of other government lawyers, you say. 70K? Starting salaries for freshly minted JD's at white shoe law firms are in the $125-$150K range. You going to pay the SEC lawyers that much? Honey, that's what the director of a government agency makes, not his enforcement staff. Meanwhile, the gap between what experienced lawyers in the public and private sector make is even more stark. And the same dichotomy holds true for accountants. You can make more setting up shop as a private accountant (after a couple hungry years, anyway) than you will ever make as a government agent. And while the FBI and the Justice Department are able to attract some talent anyway, they attract it largely because going after criminals is sexy, which going after the guy who erroneously booked his depreciation on a piece of equipment in Q1, even though the machine didn't ship until Q2, is not. And because there are lucrative opportunities in the private sector for people who put in a couple of years at the FBI or the Justice Department. The same is true of working for the SEC, of course. . . and those lucrative opportunities are located at the firms you are asking the SEC enforcers to descend on like the proverbial ton of bricks.

3) Many of the rules that need changing need Congress to change them. Congress is filled with two kinds of people: people whose eyes glaze over and whose palms begin to sweat when you start saying things like "NOL carryforwards" and "insufficient allowance for return of merchandise received"; and people who understand these terms, indeed revel in them . . . because they spent a significant amount of time working for one of the banks or law firms you want them to regulate, made most of their money that way, and still count the titans of that industry as close personal friends.

4) The public doesn't understand anything about accounting regulations. This makes them easy to demagogue. Hell, I understand accounting regulations better than, apparently, 99% of my readers, and 100% of journalists. Yet I often have to have Mindles or a similarly erudite person take me gently by the hand and explain to me why my take on a particular regulation is. . . ahem. . . idiotic. It is very, very difficult to make an informed decision about an issue like, say, the expensing of stock options.

(Thank you, incidentally, to those anonymous fellows who emailed to say "I think you're an idiot. Why not just pass a simple regulation?" It is instructive in these situations, I think, to ask yourself something: given that all the people arguing about it know quite a lot about accounting, and given that you don't know anything about accounting, do you think that it is more probable that the answer really is simple and everyone else is just an idiot, or that the answer is very complicated, and they're not the idiots here? If you answered the former, why don't you go find some physicists and lecture them on how you could do things better. The accountants already have enough back-seat drivers.)

Because the public has no realistic hope of actually making intelligent decisions on issues such as that one, agencies tend to careen between two states of affairs:

a) No one is paying attention to what they do, because it's complex and boring.

The agency regulates in a way that benefits the people who do care -- the companies, and their lawyers and banks -- because those are the people who are going to go to congress and complain if the SEC rules against them. Witness the march on congress of the technology industry in the face of a proposed change in the taxation or regulation of stock options. These are not, by and large, executives who are concerned about aggrandizing their own wealth (well. . . not solely concerned, anyway.) These are executives who are genuineley concerned about their ability to attract both capital and prime techie talent if the regulations are changed. They're self-interested, but not necessarily in a way that seeks to screw the rest of us for their own benefit. And in California, they vote. Cross them, and you can expect to have Boxer and Feinstein screaming in your ear in no time flat.

b) Something has gone horribly wrong and everyone is paying attention, and the public is screaming for you to do something now.

At this point, journalists, who know just about as much about accounting as they know about 7th Century Mayan Pottery (much less, in many cases) are going to be paging through all the regulations, searching for one that sounds relevent and can be paraphrased in a single sentence. Chances are better than even that they will either misunderstand the regulation (such as the journalists who appeared to think that the reason Enron's off-balance sheet entities were fraudulent was that they were located offshore), or totally misunderstand its implications (such as the journalists who think that banning accounting firms from consulting will eliminate conflicts of interest, without noticing that the audit firms still get all their revenue from the same source, the companies they audit; they just get less of it now.). Because they haven't the faintest clue how to read a financial statement, and aren't going to waste valuable time learning just so they can do some stupid story on financial statements, they are going to get the majority of what they report factually wrong, and come up with suggestions that will be as well thought out and useful as a journalist suggesting to an aerospace engineer that because birds can flap their arms and fly, we ought to be able to do so as well. Those will be the stories that the public reads that generate such useful thoughts as "Why is it so hard to just pass a simple regulation?"

Then their congressmen, aided by political science graduates who get all their ideas about accounting from the journalists, are going to pass some laws based on what will make the voters who read the journalists' stories think that their congresscritters are doing something about the problem. Given the public process by which the likely subject of those new laws has been selected, there are three possible outcomes:

i) The law will solve more problems than it creates.
ii) The law won't do anything much, or will create as many problems as it solves
iii) The law will create more problems than it solves.

Given the fact that the people involved have absolutely no background that would enable them to actually assess the effects of the laws they pass, the best-case scenario is that it mimics a random distribution: in other words, you have a 2/3 probability of doing nothing worthwhile. The worst-case scenario is that the bad ideas are easier to explain, and thus more likely to get passed, than the good ideas. Anyone who has ever actually tried to get the government to do something can judge for themselves which is more likely.

5) IT IS NOT POSSIBLE FOR THE SEC OR THE PRIVATE AUDITORS TO CATCH A COMPANY DETERMINED TO COMMIT FRAUD. I don't like to shout, but I'm a little fatigued by people who seem to imagine that a "beefed up" SEC is going to be some kind of real-life Hall of Justice, where accounting heroes swashbuckle about uncovering corporate fraud and abuse. First of all, as I pointed out above, you can't afford to hire the superheroes. And second of all, you have absolutely no idea about the scope of what you're proposing.

The public thinks that when Arthur Anderson went in to check a company's books, they poured over each and every entry, scrutinizing it for possible fraud, checking that it all tallied, and interrogating the executives about anything even slightly suspicious. Or rather, they think that this is what Arthur Anderson was supposed to do, and didn't. Child, when an audit team goes in, they take last year's results, check any "material variances", a.k.a. big changes in the numbers, and do a few spot-checks. That's how audit works at all teh companies. And that's how it will continue to work, because it's just not possible to apply the kind of scrutiny you're imagining.

It takes companies a fleet of accountants, payroll people, managers, and finance staff to generate all the numbers you're talking about. Checking everything they do would require an even bigger fleet. Plus, you'd have to make the number of finance, etc. employees 3 times bigger, because they'd be spending a significant portion of their time explaining to the auditors what they'd done. Which would in turn require more auditors. . . meanwhile, the company can't actually make or sell anything, because they're too busy explaining what they made or sold last year to the auditors. Compared to the level of scrutiny currently in place, you're talking about a simply massive reallocation of resources to audit.

Say getting a really good audit requires increasing the number of employees by 1%. That's probably conservative. You're talking about taking enough workers out of the limited pool of skilled labor and transferring them to do work that is economically unproductive: all it does is check that companies actually made the stuff they say they made. Now, it's worth it if the damage to the economy is more than, say, 1% of GDP. But that isn't the case. The scandals were bad, and they hammered the market. But the market was going to get hammered by something, because it was enormously overvalued. The cost is all out of proportion to the benefit.

Meanwhile, if the company really, really wants to lie, they're going to put in five of their really sharp folks to surround the area they're lying about, and you're not going to catch them, because they're good liars, and because they know what you're looking for, but you don't.

Increasing the SEC budget may be worthwhile, but absent the kinds of structural changes Mindles is talking about, it's not going to fix the problem. And those changes can only come from reasoned debate between people who understand the issues, not mass hysteria.

Posted by Jane Galt at October 23, 2002 10:13 AM | TrackBack | Technorati inbound links