Hello, all. The last time I guest-blogged here, my first post was a rather unpersuasive attempt to develop a fixed-cost/variable-cost theory of reasonable attorneys’ fees in the course of defending a Seventh-Circuit opinion by Judge Posner in Budget Rent-A-Car v. Consolidated Equity, where Posner reprimanded a party for demanding excessive attorneys fees. Coincidentally, another fascinating question about the costs of doing legal business came from the Seventh Circuit today, this time by Judge Easterbrook in Szopa v. U.S.. After deciding that a tax protester’s repeat litigation is frivolous, Easterbrook agrees to award sanctions, but questions the government for demanding sanctions several times higher than previous ones have been (adjusted for inflation). The government claims to spend $11,000-$12,000 on a frivolous claim, but after some back-of-the-envelope math, the court concludes that if the government really is spending that much money on each claim, it is probably doing something wrong. I will spare readers the attempt at cute economic just-so story this time, but I am confused about two tensions in the court’s sanctions ruling:
On the one hand, the court and the government both clearly see a relationship between the cost of refuting a frivolous appeal and the sanction imposed on the frivolous litigant. This makes sense; those who abuse the legal system should pay the costs of abusing it. On the other hand, the court doubles the sanction because Szopa is a recalcitrant recidivist tax offender who has proven difficult to deter with sanctions. But if the sanction really is equal to the cost of the frivolous lawsuit, then why should we care if the offender is deterred? Punitive damages make sense only if she either is not paying the full social price of her conduct, or if her conduct is hard to detect and punish. One possibility is that Szopa visits costs on the system at large– like taking up Judge Easterbrook’s (and his law clerks’) time in writing the opinion– not included in the government’s litigation costs. But if so, shouldn’t those costs be part of the sanctions calculation in the first place? Another possibility is that frivolous-tax-protester lawsuits are not always “caught” and therefore the punishment needs to be higher to make up for the imperfect enforcement rate. But this seems unlikely. The punitive sanctions are imposed only for making frivolous arguments in court. Making frivolous arguments in court has an incredibly high enforcement rate since, after all, there is a judge (or three) sitting right there. So it seems unlikely that the court believes that Szopa is secretly making twice as many frivolous filings, and being caught only half the time. So, what gives? Does $2722.58 actually fine the litigant for the full cost of her conduct? If not, why not charge Szopa the full social cost? If so, then doesn’t doubling the sanction overdeter?
Posted by Will Baude on July 5, 2006 at 05:00 PM
» Still doing business from Crescat Sententia In two previous bouts of guest-posting, I discussed some interesting Seventh Circuit cases about the appropriate sanctions and fees for a frivolous appeal. In my first post, when I defended Judge Posner’s opinion in Budget Rent-a-Car, some commenters s… [Read More]
Tracked on Nov 4, 2006 8:48:37 PM
Comments
Will,
I don’t think that either sanctions under Code Section 6673 (even if uncapped by Congressional action) or sanctions under Rule 11 and/or the section of Title 28 that governs vexatious litigation (1927?) could ever get at the full scope of the tax protestor problem, because those can only be imposed with respect to behavior that takes place within the judicial system. There is a lot that goes on before a tax case gets to court. The Saltzman treatise on IRS practice and procedure is the indispensable resource on all of this stuff.
Short version:
Before a typical tax case gets to court, (1) the return (assuming a return was filed — non-filer cases work differently) was screened both by a computer and live people to determine that it had “audit potential,” (2) one or more IRS examination people spent time (in some cases, thousands of hours) obtaining and reviewing information from the taxpayer, identifying legal issues, and trying to resolve them informally, (3) if the case doesn’t settle in Exam, the taxpayer has the option to pursue an administrative appeal within the IRS (which happens more often than not, for a variety of reasons including that exhaustion of administrative remedies is a prerequisite for an award of attorneys fees against the IRS under Code Section 7430).
Keep in mind that protestors love to gum up the system, so their cases often consume greater-than-average amounts of time and resources in the examination and appeals processes.
The total time charged to a case involving a sophisticated protestor, prior to the filing of a petition in the Tax Court or a complaint in one of the refund forums (IRS employees keep timesheets), is almost certain to be deep into the hundreds of hours, if not into the thousands. Now, most of that time is logged by IRS employees at grades GS-13 or below, and most IRS litigators are at grades GS-13 or GS-14 (GS-14 currently tops out at about $110K/yr) so the prorated direct costs are not astronomical in any particular case. But there are thousands upon thousands of such cases every year. And in the eyes of IRS leadership, all of them have to be fought tooth and nail in order to protect our system of voluntary compliance.
One final point: IRS litigators probably pursue section 6673 sanctions in a less than optimal number of cases, because there are significant institutional barriers to obtaining permission to engage in what are seen as “aggressive litigation tactics,” especially against pro se litigants.
HTH, at least a little.
Posted by: Totally Confused | Jul 8, 2006 2:12:04 PM
Frivolous claims generate benefits for law nerds (a term I use lovingly). Did not his claim generate benefits for everyone who enjoyed reading, or participating, in this thread?
Posted by: washerdreyer | Jul 6, 2006 7:50:00 PM
Yes, I should have been more careful, and limited my claim to the context of these cases. At the appeals stage (which I think is all that Easterbrook is assessing sanctions for) does the government ever settle tax protester claims?
Posted by: William Baude | Jul 6, 2006 2:27:47 PM
Making frivolous arguments in court has an incredibly high enforcement rate since, after all, there is a judge (or three) sitting right there. Not always. Some cases settle, or are withdrawn, but only after legal costs have been incurred.
Posted by: Bruce | Jul 6, 2006 2:11:12 PM
TC,
This is fascinating, and I wonder if you’d be willing to elaborate the costs a bit for some of the readers who are (like me) less intimately familiar with the procedures involved.
For example, would taking into account the pro-rated salary of the lawyers involved with the case (and the IRS officials prosecuting it) therefore also account for all opportunity costs? About how many hours does it take to deal with a frivolous tax protester. Easterbrook thought it unreasonable that a response to a frivolous appeal would take ten full days of senior legal work. Is he right?
Posted by: William Baude | Jul 6, 2006 11:35:46 AM
The idea of imposing cost-internalizing monetary sanctions appears to have some merit. While it would probably not reduce the number of frivolous tax claims to zero, it seems likely to substantially reduce them.
However, as far as the Tax Court is concerned, it would require legislation. Internal Revenue Code Section 6673 caps the amount of sanctions the Tax Court can impose in such cases at $25,000. Speaking from unpleasant personal experience, I can assert with some confidence that under any reasonable market measure of the value of an experienced tax litigator’s time, the direct cost to the Office of Chief Counsel of the IRS of defending against one of these cases exceeds (often substantially) $25,000. And that doesn’t even begin to compensate the taxpaying public for the opportunity costs incurred when scarce IRS resources that could be applied to litigating cases involving real and substantial legal issues have to be reallocated to dealing with tax protestors.
The real villain of this piece, when you get right down to it, is the Supreme Court’s breathtakingly misguided opinion in Cheek, which made it virtually impossible to successfully prosecute tax protestors for evasion. A good-faith but objectively unreasonable belief that one is not subject to tax seems to be fairly easy to summon up if one is of a certain world-view.
Just my 3.3 cents worth (that being the amount of pre-tax income that, when taxed at the maximum Federal rate, leaves two cents left over).
Posted by: Totally Confused | Jul 6, 2006 11:25:43 AM
Dave,
This is possible. The idea would go that the optimal number of frivolous lawsuits is zero, since they have zero societal benefit, so we actually want to stamp them out entirely. Of course, courts don’t assess sanctions in this way: 1, we foreclose access to the litigation system only after several offenses, not the first. 2, we don’t charge million-dollar sanctions. Now perhaps we don’t do these things because of the possibility of erroneously punishing a worthwhile claim, basic mercy, etc.
But if this theory of the optimal number of frivolous lawsuits is right, then Easterbrook’s opinion is wrong. If the only things that keep us away from punitively-high sanctions are “sympathy” and “effective”ness, then why would Easterbrook reduce the sanctions in this case from the ones the government has asked for (and gotten elsewhere)? And why would Easterbrook think it was relevant what cost the behavior imposed on the system as a whole?
Put differently: You can try to set the sanctions such that they internalize the costs of the sanctioned behavior. Or you can try to set sanctions such that you minimize the amount of the sanctioned behavior (subject to some constraint, like mercy). But you can’t try to do both things at the same time, because they have different economic and mathematical implications.
The cost-internalizing sanction and the frivolous-litigation-minimizing sanction are not the same, precisely because the costs of frivolous litigation are not infinite, so if you merely internalize the costs you get non-zero equilibrium frivolity. And if you attempt to stamp out all frivolity, you are doing more than simply internalizing the costs.
Posted by: Will Baude | Jul 6, 2006 11:05:46 AM
Will,
Well, I think the answer to your point is that we don’t really care about the benefits to the filer of frivolous lawsuits. Such lawsuits produce no benefit to society, and they do produce tangible costs, so we should stamp them out, however we can. I think the only reason you don’t charge a guy like this, say, a million dollars, or whatever he’s worth, is that (a) we have some sympathy for a crackpot, and (b) there are more effective ways to do it, as Totally Confused pointed out.
Posted by: Dave | Jul 6, 2006 10:56:46 AM
TC,
Note that Easterbrook does in fact order the Seventh circuit to stop accepting filings from Szopa if the sanction goes unpaid.
The optimal rate of frivolous filings is non-zero because frivolous filings have both costs and benefits. The costs– wasted time and resources– are obvious. The benefits are less obvious, but they obviously exist, or Szopa wouldn’t file. So the simple answer is that if Szopa is happy to file, even after paying the cost of frivolity, the psychic benefits to the filer must outweigh the costs to society. So why stamp them out?
Posted by: Will Baude | Jul 6, 2006 7:15:45 AM
As a former IRS litigator, I will freely admit to vicariously having a dog in this fight. People like this are an absolute nightmare to deal with; they nearly always proceed pro se, and are typically possessed of some …aaaahhhh … curious notions about how the judicial system works.
However, my personal animus against tax protestors aside, I am struggling to understand why the optimum level of frivolous lawsuits would be anything other than zero. Perhpas someone could enlighten a tax lawyer of very little brain.
An alternative approach to dealing with such folks, which moots the issue about the deterrent power of monetary sanctions, is to instruct the clerk of the court not to accept for filing any future complaints that allege bad acts by the Government in connection with the assessment or collection of taxes. It seems to me that the authority to enter such orders can be found in some combination of Rule 1 of the Federal Rules of Civil Procedure and the court’s inherent authority to control its own docket.
Posted by: Totally Confused | Jul 6, 2006 12:57:17 AM
Dave,
I guess one of us is begging the question of what it is to overdeter. If an action has both costs and benefits, and you force the actor to pay for all of the costs, then why should we care if she does it, which she presumably will only when the benefits are larger?
Put differently, who is to say that sanctions haven’t “worked”? That’s true only if you assume that the optimum level of frivolous filing is zero. Which we obviously don’t.
Posted by: Will Baude | Jul 5, 2006 11:24:35 PM
Thnaks for this post. I would love to hear more about what you sense would resolve the on-going problem you raised so well here. I for one, would be interested in your solution… and sense you have a great one… just from what I read here. Very interesting…
Brain Based Business
Posted by: Ellen Weber | Jul 5, 2006 11:08:18 PM
Why should we worry about overdeterrence here? Precisely because there is perfect enforcement (i.e. courts have the offense right in front of them, and know it when they see it), then I don’t see why we shouldn’t try to overdeter as much as possible. We know the behavior exists; we know previous sanctions have not worked with this particular offender; and there are no adverse effects from having a chilling effect on frivolous lawsuits: unless you’re also someone who files frivolous lawsuits, in which case we don’t worry about chilling your behavior.
Posted by: Dave | Jul 5, 2006 8:28:03 PM