It’s sweeps week for corporate crooks.

On Monday US District Judge Leonard Sand sentenced John Rigas, the ailing octogenarian founder and former CEO of Adelphia Communications, to fifteen years in federal prison. Rigas was convicted last July of fraud and conspiracy charges relating to $50 million in cash advances and $252 million more in margin loans. He has suffered from bladder cancer and underwent triple bypass surgery several years ago.

Rigas’s sentence should be of some concern to L. Dennis Kozlowski, the former CEO of Tyco, and Mark Swartz, his main lieutenant. Both of them were convicted by a Manhattan jury just three days earlier on charges relating to the theft of $150 million and the covert sale of stock worth nearly $500 million more. The Tyco convictions (twenty-two counts each) cap a three-year investigation involving two trials, nearly ten months of testimony and several weeks of jury deliberations. Swartz and Kozlowski’s sentencing hearing is scheduled for August 2.

But as huge as a half-billion-dollar fraud may be, both the Adelphia and Tyco cases pale in comparison to the verdict returned on March 15 against former WorldCom CEO Bernard Ebbers, who was convicted of engineering an unprecedented $11 billion fraud. Ebbers is set to be sentenced in a few weeks.

Because the Supreme Court has declared the federal sentencing guidelines to be advisory rather than mandatory, US District Judge Barbara Jones (like Judge Sand before her) will have extraordinary latitude when she sentences Ebbers next month. Technically free to impose a sentence she deems just in light of all the facts and circumstances, Judge Jones could, in theory, order anything from home confinement to a functional life sentence (Ebbers is 63). And she is not alone. Even in state court, where Kozlowski and Swartz will meet their fate, Judge Michael Obus has the discretion to impose a sentence that could wind up being as short as a year or as long as two decades.

Both judges are allowed to do something else, too: Each may make a nonbinding recommendation about where the defendants should serve their respective sentences. (The Bureau of Prisons makes the final decision.) And though the headlines are certain to focus on the number of years each judge imposes, it is actually this recommendation, rather than the length of the sentence, that may be the most effective way for a judge to deter corporate crime.

Most sentencing schemes rest on two philosophical pillars–the idea of just deserts and the notion that penal sanction decreases the likelihood of future crime, either through the incapacitation of the offender or by deterring others. And while just deserts are often invoked in the stern lectures of judges sentencing individual offenders, it is the power of deterrence that is regularly cited by politicians around the country to justify ever-harsher criminal laws and ever more draconian sentencing measures. Standing before the cameras, our lawmakers insistently talk of insuring that bad guys will “think twice” before committing whatever act they happen to be railing against.

But will they?

Deterrence is a funny thing. In a 2001 paper that surveyed the research on the deterrent effect of the death penalty, the Maryland State Commission on Criminal Sentencing Policy concluded that the death penalty does not have a deterrent effect. Similar studies in different contexts have come to the same conclusion. To many people who have spent time in the system, these findings make sense–after all, the idea of deterrence rests on the assumption that criminal defendants are rational actors dispassionately assessing criminality as a life choice.

In fact, though, very few of those processed by the criminal justice system are actually rational actors–instead, most are driven by impulses far more powerful than reason. To a crack addict, the only question is whether a robbery will net the $5 necessary for another hit. Two years or 200 is utterly beside the point. For most of the indigent involved in the criminal justice system, crime–whether robbing someone to get high, beating or shooting someone in a fight or fit of rage, or stealing out of hunger–is about addiction, fury, fear, hunger or need. They don’t stop to wonder about punishment, and no amount of penal law posturing will stay their hands.

Ebbers, Kozlowski and Rigas, on the other hand, represent something we rarely see in the criminal justice system: true rational actors.

Ebbers and his ilk are intimately acquainted with the nuances of criminal law, the boundaries of accounting fraud and the implications of measures like the Sarbanes-Oxley Act, which is designed to make them more accountable. Moreover, they have access to expensive accountants and advisers well versed in the enforcement practices of the regulators who oversee their industries. Men like Ebbers (or Ken Lay or Jeff Skilling, who have yet to face trial in the Enron debacle) weigh their options–objectively assessing risk and reward. This may make them canny crooks, but it also makes them supremely responsive to the deterrent factors that most legal economists wrongly imagine apply to everyone–chief among them is fear.

Traditionally, there is not a lot of fear in the tony precincts where white-collar criminals dwell. The truth is, we rarely send white-collar offenders to serve hard time. Because of their wealth, privilege and relative sophistication, white-collar criminals can count on gaining significant advantages at every stage of a criminal justice proceeding.

Because of spotty enforcement, white-collar criminals are far more likely to get away with their crimes than poor folks. And when they are caught, wealthy corporate executives can take refuge in their powerful friends and associates while availing themselves of high-priced lawyers, jury-selection experts and mitigation specialists. The truth is, most corporate crooks get the best representation money can buy, and money, in the criminal justice system can buy quite a bit. Unlike average criminal defendants, who have few reasonable expectations about where they will do their time, white-collar criminals usually employ consultants to help insure that the defendants serve their sentences under conditions of confinement that, while never pleasant, are nonetheless tolerable. And it works. Because so much of corporate crime is adjudicated in federal rather than state court, corporate criminals often serve their time under relatively cushy conditions of confinement.

Indeed, a slew of high-profile defendants, from Alfred Taubman to Martha Stewart, have recently emerged from their sentences trim, posh and still profitable. And though extended confinement like that initially imposed on junk-bond king Michael Milken may be something of a deterrent, relative rarity and palatable conditions of confinement severely limit the deterrent effect on those considering multimillion-dollar schemes that run afoul of the law. Put bluntly, it’s not irrational to steal $10 million if the worst-case scenario is a few years in Camp Fed. But change that sentence to read Sing Sing or Attica or Pelican Bay and what emerges is a whole new calculus of crime.

In dealing with rational actors, it may well be that the conditions of confinement matter far more than the length of the sentence.

White-collar crimes are about greed and self-aggrandizement, and while those things may be deeply compelling, they are fundamentally different from hunger, addiction or desperation. Unlike them, greed responds to fear, and fear is more about the conditions of confinement than its duration. The truth is that to most people, the prospect of even a short stint in a maximum-security prison is far more frightening than years in Camp Fed.

So if during the upcoming parade of CEO sentencing Judge Jones wants to “send a message” in order to deter corporate crime, maybe she should suggest that the Bureau of Prisons save a cell in Leavenworth for Ebbers and his ilk.