Saturday, October 24, 2015

A Couple of Obvious, and One Not-so-Obvious, Responses to Judge Emmett Sullivan's Question About Deferred Prosecution Agreements

Thursday night, I was at a reception for alumni of my old firm and ran into a former colleague who now spends his time as in-house counsel to a large foreign firm that is the subject of widely publicized federal and state criminal and other investigations.  I asked him how the job was going and he said, in substance, "We have over 10,000 people working for us, but these investigations are the result of the actions of 5 or 6 of them,  Still the whole organization is being punished for what they did".

I immediately thought of that conversation when I read reports yesterday that U.S. District Judge Emmett Sullivan in Washington D.C. had issued an 84-page opinion (which can be downloaded at the preceding link) musing on what he perceives to be an unwise disparity in the application of criminal law to corporations and individuals.  Specifically, Judge Sullivan complains that prosecutors are willing to use "deferred prosecution agreements" ("DPAs") in dealing with corporations but should offer the same opportunity to individuals, which he believes was the original intent for inserting the DPA mechanism as an exception within the "Speedy Trial Act".  His concerns are pure dicta as his opinion grants the motion of the Department of Justice for approval of DPAs for two privately held corporations alleged to have committed bribery in the pursuit of federal contracts.  Still, they are catnip to the likes of the New York Times which has persistently displayed an embarrassing lack of understanding of the basic aspects of corporations and so I thought I would write a more informed summary of Judge Sullivan's decision, which is far from the raving left-wing screed the Times article implies it to be.

In one of the cases, four individuals -- two employees of the Army Corps of Engineers and two employees of the company seeking the contract -- were prosecuted and made guilty pleas.  In the second, which involved bribes sought and paid in South Korea in relation to contracts pertaining to the U.S. Armed Forces' presence there, no individuals have been prosecuted yet, but that DPA requires the executives of the defendant company to cooperate with the government in its ongoing investigation.

Judge Sullivan first reviews the legislative history that gave rise to the DPA exception under the Speedy Trial Act deadlines.  He concludes from it that federal district courts are intended to give limited review to DPAs.  The section of the Speedy Trial Act that authorizes DPAs (28 U.S.C. 3161(h)) explicitly requires court approval, but that requirement is lodged in a sentence that says the DPA must be entered into "for the purpose of allowing the defendant to demonstrate his good conduct".  Thus, Judge Sullivan  determines judicial scrutiny is limited to ensuring that the DPA is really about "diversion" of the defendant into a supervised program, and not an attempt to endrun other aspects of the Speedy Trial Act, citing S. Rep. No. 93-1021 at 37 (1974).  He also allows that the district court would have the inherent power to withhold approval over "especially problematic" DPAs, giving examples, drawn from United States v. HSBC Bank USA, No. 12-cr-763, 2013 WL 3306161 (E.D.N.Y. July 1, 2013), such as provisions conferring personal benefits on the prosecutor or on persons or institutions of interest to the prosecutor.

His conclusion appears to be closer to the position the government argued before him and in the HSBC case than to the broader view of the judge's role taken by Judge Richard Leon, in United States v.Fokker Servs., B.V.,79 F.Supp.3d 160 (D.D.C. 2015), appeals docketed, Nos. 15-3016, 15-3017 (D.C. Cir. filed Feb. 23, 2015).  In Fokker, Judge Leon became apparently the only federal judge in history to reject a DPA because it wasn't harsh enough relative to the alleged conduct (violating export control laws relative to Iran for over 5 years (recall Fokker is a non-US-based company)).  Without explicitly disagreeing with Judge Leon, Judge Sullivan expresses significant separation-of-powers reservations about the judiciary usurping the executive branch's role in determining whether to prosecute someone, individual or corporate.  (Sorry New York Times!)

The judge then subjects the two agreements before him to said limited review and approves them,  Although it's not my focus here, it's at least relevant to spend a minute on the backstory to gain some context for why prosecutors might use DPAs for these two cases.  In the case of the Army Corps of Engineers bribery, recall that it was the government employee who requested the bribe.  It's not exactly entrapment, since the employee wasn't a law enforcement officer, but it makes criminal sanctions a little less compelling since there isn't as much deterrence value in punishing the second to act in a criminal transaction. Also, assuming that a conviction would disqualify the private company from further Army Corps of Engineers work, it seems a little disingenuous to punish the company for conduct solicited by someone at the Army Corps of Engineers.

Somewhat similarly in the case of the South Korean contract bribery case, the defendant who entered into the DPA was a subcontractor who, again, was solicited by the person working for the Army in administering the contract so some of the same factors about punishing the second to act again come into play.

Judge Sullivan then appends the dicta that caught the attention of the Times.  He states that the legislative authority for DPAs was created because the Senate, circa 1974, was impressed by two experimental "diversion" programs implemented in the late 1960s (!) in New York City and Washington D.C., which, he observes, only processed individuals who were not accused on homicide, rape, kidnapping or arson (other criteria as well).  (Apparently persons who committed armed robbery or breaking and entering or sold heroin were eligible for diversion....).  The programs provided "counseling" and "employment" and were supposedly successful as of 1974 (having lived in NYC during the 1970s, I have to say I find that conclusion utterly ridiculous).  Also, it is hard to buy into someone endorsing a New York City "diversion" program in the wake of the murder of Randolph Holder, a New York City police officer, by Tyrone Howard, a career criminal who had been placed in a "diversion" program just this past December after an arrest for selling crack. Nevertheless, I'll try to give an accurate summary of Judge Sullivan's call for expanding the use of DPAs for individuals.

First, he contrasts (a) the use of DPAs for individuals, which amount to a very small portion of all DOJ decisions not to prosecute with (b) the rough equivalence of corporate DPAs to corporate prosecutions.  He notes that the statute does not suggest the use of DPAs for corporations is unwarranted, but rather that they should be offered more often to individuals.  He accompanies this contrast with a complaint about the use of a DPA in regard to GM's ignition switch conduct.  He writes "people are no less prone to rehabilitation than corporations. Drug conspiracy defendants are no less deserving of a second-chance than bribery conspiracy defendants. And society is harmed at least as much by the devastating effect that felony convictions have on the lives of
its citizens as it is by the effect of criminal convictions on corporations."  He cites no authority for these beliefs.

Although I agree with the judge that the failure to prosecute culpable individuals at GM is troubling, his overall position shows, I am afraid, that the judge does not very much about corporations. You can't equate a person and a corporation in this fashion.  As illustrated by my former partner's remarks above, when a large corporation does something wrong, it is usually a very small fraction of its employees that are involved in that wrongdoing. A corporation is typically packed with innocent, law-abiding employees.  In contrast, when an individual does something wrong, 100% of that individual committed the crime. There is no option to separate the part of the individual that committed the crime from some part that didn't.  So it strikes me as fallacious and naive to say that people are as likely to rehabilitate as corporations. When you remove the people who act criminally from a corporation, you are going to be much more likely not to see the corporation repeat that kind of conduct again. In the case of GM, virtually everyone involved at GM today, from workers to shareholders had no involvement in the ignition-switch liability and can rehabilitate the corporation quite easily just by continuing to act as they did, once the culpable individuals are excised.  You can't excise a small part of an individual and rehabilitate the rest.

Let me offer a further example relevant to the proposition of prosecuting corporations for acts of their non-control-group employees.  Recall that in the bribery case at the Army Corps of Engineers, the bribe was solicited by an employee at the Corps.  Yet no one even dreams of saying the Army Corps of Engineers should be convicted of a crime.  Why not? It's the same "respondeat superior" argument. And frankly, this is definitely not the only time an Army Corps of Engineers employee solicited a bribe.  I plugged "Army Corps of Engineers bribery scandal" into Google and the first page had FBI and newspaper reports of bribery arrests and investigations involving Army Corps of Engineers contracts from every year this decade.  So why shouldn't the Army Corps of Engineers be treated as a recidivist bribe-seeking organization and convicted, the way progressives call for corporations to be convicted?  Recall that many of the claims bought against banks by the Obama Administration were based on an interpretation of FIRREA that made it a crime for a bank to harm itself.  So if that is a valid basis for criminal charges against private sector institutions, why not for public sector institutions?

Of course, the answer is, the bribe seekers were not carrying out the Army Corps of Engineers' mission, they were off on a "frolic and detour", etc.  "Respondeat superior" does not come into play, you can't attribute the individuals' conduct to the public-spirited Army Corps of Engineers.  Second, what good would it do for the citizens to convict a federal organization? Would it have to go out of business and how would that be in the public interest?  All of those may be legitimate observations or concerns, but they are identical in the case of a privately held corporation who receives a bribe request from an Army Corps employee. Private sector and public sector organizations should be treated alike, it seems to me, in terms of criminal exposure for acts of their employees.

Judge Sullivan goes on to recite a number of proposals / efforts to reduce incarceration in federal prisons particularly for drug offenses.  He lauds these and talks about the potential for responses to drug dealing other than incarceration to result in a net increase in welfare to society. He cites no data. But quotes President Obama that we are "a nation of second chances". Personally, although I think judges should have greater discretion over sentencing, I am skeptical about the claim of efficacy of diversion programs, and not just because of the Randolph Holder killing.

Skeptical because, although prison is really bad for people, it's naive to think that they will rehabilitate-in-place, in the same environment they were committing crimes in. It's the classic fallacy of policy debate to say X is bad, so we should do Y instead, without showing Y is better. It's easily possible that X and Y are both bad, so which one is worse and for whom - that's the real policy question.  That a person committing a crime is privileged to receive governmental assistance over a victim is a difficult proposition to sustain.  If you don't want to put people in jail for some action, let it be legal and get rid of the collateral damage that way. But if it's a crime, returning people to the environment of the crime strikes me as likely to have as high a rate of failure as incarceration, unless magically you get all the criminals in the area into a rehabilitation program on the same day.

More significantly, I think a lot of drug arrests in urban areas are of people who the cops believe are committing other, more violent crimes, about which  no one is willing to testify for fear of violent reprisals.  These arrests are a modern version of the arrest of Al Capone for federal income tax evasion -- a crime that the government can prosecute without a "snitch" because you only need a police officer to testify to possession. The arrest record of the man who killed Randolph Holder is a perfect example of this. He had been arrested 23 times but had no conviction for violent offenses. In few  cases, I suspect, we are talking about true "second" chances. Tyrone Howard may be an extreme example, but 23?

But Judge Sullivan doesn't even touch on the real motivation for corporate DPAs vs individual DPAs.  With corporate DPAs, prosecutors make money.

First, DPAs often involve a fine being paid.  And often the prosecuting body keeps a good chunk of that fine for its own operating expenses. Per The Economist from August of last year:  Contrary to the conventional wisdom,” write Margaret Lemos and Max Minzner in an article in January’s Harvard Law Review, “public enforcers often seek large monetary awards for self-interested reasons divorced from the public interest in deterrence. The incentives are strongest when enforcement agencies are permitted to retain all or some of the proceeds of enforcement—an institutional arrangement that is common at the state level and beginning to crop up in federal law.” (The full-text of the HLR article is here.)

Second, corporate DPAs have intensive monitoring, typically done by a private lawyer who used to be -- shocking, I know -- an ex-prosecutor, whose fees are paid by the company party to the DPA.  So prosecutors have a strong interest in the continuance of DPAs because they can envision, when they are ex-prosecutors, getting paid to be monitors!  Men and women who made maybe $60,000 in a year can collect more than that in just a month of billing at $700 per hour and up.  That's the real reason why corporate DPAs have spiked in recent years.  Whereas individual DPAs just cost the fisc money to carry out and supervise, corporate DPAs get private sector money to flow into the prosecutors' budgets and provide good income for ex-prosecutors once they leave office. It seems to me the power to scrutinize DPAs for provisions that are "especially problematic" ought to be focused on these monitoring arrangements. It seems to me pretty sleazy that criminal liability is avoided by a mechanism that results in six, seven and eight figure payments to private sector actors.  How is that not "especially problematic"?  If a company is going to pay that kind of money and avoid criminal liability, it would seem much cleaner to have the "monitoring" done by the FBI, the GAO, or some other government watchdog, and have the payments fixed and approved by the court at the time the DPA is approved, and paid to the fisc.

At the end of the day, though, in my opinion, corporations shouldn't be prosecuted unless the criminal behavior runs all they way up to the C-suite and board level.  That doesn't rule out restitutions, regulatory fines, civil liability, etc. It's just that criminal sanctions are misplaced when it comes to a legal person that has no will of its own, just a large number of employees the vast majority of whom are not culpable in any manner.  In this light, corporate DPAs would by and large go away and the utility of DPAs for individuals would be assessed, as it should, without the use of spurious analogies.

Tuesday, October 20, 2015

An Appellate Court Shows How To Analyze Escrow Accounts in Bankruptcy

One of the recurring frustrations of my practicing years was having to confront a widespread ineptitude among lenders, borrowers and their counsel concerning how to set up an escrow account to protect its contents from being siphoned into the future bankruptcy of the counterparty to the escrow.  Up to and even well after the decision of the Second Circuit in In re Vienna Park Properties, 976 F.2d 106 (2d Cir. 1992), which avoided a lien on a poorly structured escrow account, I frequently encountered lenders and their counsel, whether in-house or outside, who would frequently just throw money into an account at some "escrow agent" and walk away from the closing thinking they had "perfected liens" on the money in the "escrow account" which meant they could just take it upon a default or bankruptcy.   This was a particular problem when said ill-advised lender would come to me in another deal with a term sheet calling for such an arrangement and then be perplexed and suspicious when I would explain to them that their request for a "perfection opinion" from my firm on their "lien" on the money in the escrow account was going to be more harmful to their interests than helpful and the best thing I could do to protect their interests was to restructure the terms of the arrangement.  It was a practical lesson in the lawyering version of "Gresham's Law" (that bad money drives out good) -- in transactional lawyering, it is often the case that, if bad lawyering gets somewhere first, a client sometimes might just as soon prefer not to learn about the risk it's holding in other deals.

So I was gratified to see in Tuesday's Daily Bankruptcy News a New Case squib about a Tenth Circuit case decided Monday that seems to have gotten the analysis of an escrow account in bankruptcy right for a change.  When I clicked on the link to the opinion itself, I saw the opinion is not intended for publication and not to be cited as precedent, and I decided it would be beneficial to write a short post about why it is right so that its analysis, being correct, is better preserved and propagated, which in turn will lead to greater accuracy in future litigation over, and structuring of, these arrangements.

The Tenth Circuit decision is captioned In re Expert South Tulsa LLC, Case No. 15-3000 (10th Cir. Oct 19, 2015); it affirms a reported BAP opinion (522 B.R. 634, which I haven't read).  As succinctly stated in the opinion, one party to the escrow, LTF, bought a piece of land from the other party, debtor Expert South Tulsa.  In the purchase agreement, Expert South Tulsa agreed to make improvements to the land.  In an arrangement that is very common in small business and middle market commercial matters (which is why it's important that courts gets these analyses right), LTF required Expert South Tulsa to put the funds for completion of the improvements into escrow (a more costly mechanism to accomplish the same result would have been a surety bond, although the bonding company might well have required the same arrangement and certainly would have charged a fee for its involvement).  Critically, the opinion reports, Expert South Tulsa "could recover portions of the escrowed funds each time it completed a segment" of the improvements.  Of course, Expert South Tulsa filed chapter 11 before completing much of the work.  LTF initiated an adversary proceeding declaring that the funds in escrow were not part of the bankruptcy estate.  Expert South Tulsa disagreed, seemingly finding the proposition so ludicrously obvious that, the opinion notes, it didn't offer the bankruptcy judge much more than a conclusory snort that, before the money went into escrow, it was in the debtor's pocket, so self-evidently the funds were in the  debtor's estate.

The panel correctly noted that the legal characterization of a debtor's interest in property is strictly one of state law (Butner; Whiting Pools).  And under Oklahoma law, as under the laws I dealt with in my own practice, title to the funds in an escrow account belongs to the escrow agent.  What the principals under an escrow agreement have is a contractual (sometimes called "contingent equitable") right to delivery of the funds upon satisfaction of the conditions specified in the agreement.  That interest, not an interest in the funds themselves, passes to the bankruptcy estate of a party to the agreement that goes into bankruptcy.  That interest, by the way, is a "general intangible" for UCC purposes and a creditor may be wise to file a UCC against that interest to be perfected in the value it represents.  But the lender, or other counterparty to the escrow agreement should not characterize its position as having a "lien" on the escrow account itself, nor on the money in it, or a court may view that as evidence that the arrangement is not a "true escrow" but merely a "disguised cash collateral arrangement" in which case the funds will be deemed property of the bankruptcy estate and the lender will have, at best, a claim for adequate perfection and at worst, an unperfected and avoidable lien on the money in the account, as was the result in Vienna Park.

So, how does one know one has a "true escrow" and not a "disguised cash collateral" setup.  This is where the terms of the escrow agreement are fundamental.  In Expert South Tulsa, they had a true escrow because (a) the conditions for disbursement from the escrow were objective and beyond the discretion of the debtor -- it could receive funds only when it completed a segment of the improvements, an objectively verifiable situation beyond its discretion -- and (b) disbursement from the escrow did not reduce the debtor's obligation to the counterparty to the escrow; rather, it merely repaid the debtor for work previously performed.  So it looked like a "true escrow".

This contrasts with the "escrow account" in Vienna Park where (a) the debtor, through a manager of its choosing, had discretion over spending the funds in the escrow account and (b) the lender took a lien on the funds in the funds in the escrow account and the debtor's "residual" interest in those funds and the "escrow" terminated upon "satisfaction" of the debtor's obligations to the lender.  As the lender had failed to file a UCC financing statement covering those security interests (thinking they were perfected through possession even though the funds were on deposit in a third-party bank and weren't tangible in the first place), their lien was avoided and the funds were free for the estate to use in the chapter 11 case (today, that arrangement could be perfected through a "control agreement" with that third-party bank).  So the two keys to setting up an escrow to be outside a bankruptcy estate are: (1) have disbursement be based on objective criteria and not under the control of the debtor and (2) have disbursement not reduce the debtor's debt.

In the Tenth Circuit case, LTF, the counterparty to the escrow, set it up and documented it just right to keep it out of the bankruptcy estate altogether.  Disbursement was objective and not in the debtor's control and disbursement did not benefit LTF by reducing the debtor's debt to LTF; rather, it benefited the estate.  And those are the two fundamental principles that all good-against-bankruptcy escrow arrangements depend on.  Of course, as the issue is one of state law, in any given specific situation, the relevant state law may prescribe additional bells and whistles to structure the escrow in the optimum manner and practitioners and litigators should inform themselves fully about the escrow law of the relevant state before sallying forth to advise clients or advocate to bankruptcy judges.