The Senate Judiciary Committee Overhauls the Injunction Bond Bill

In a previous post, I highlighted the House Judiciary Committee’s proposal to eliminate contempt enforcement for any injunctions given without an injunction bond. The most dramatic problem with the bill was that it would eliminate all contempt enforcement for final injunctions, whether past, present or future, and whether in litigation against the federal government, litigation brought by the federal government, or litigation between private parties. In effect, every federal injunction would have instantly become a declaratory judgment—a proposition beyond the wildest dreams of Edwin Borchard, who in the first half of the twentieth century was Mr. Declaratory Judgment. The House version also raised a constitutional issue about retroactive immunization of disobedience to an injunction. And the House version could be easily evaded for preliminary injunctions and temporary restraining orders by federal courts; all they had to do was require the posting of a $1 bond. (That’s not hypothetical: Judge Breyer’s PI about the California National Guard required a nominal bond of $100.)

Enter the proposal of the Senate Judiciary Committee (h/t Hailey Fuchs of Politico):

SEC. 203. RESTRICTION ON ENFORCEMENT.

No court of the United States may issue a preliminary injunction or temporary restraining order against the Federal Government (other than a preliminary injunction or temporary restraining order issued in a case proceeding under title 11, United States Code) if no security is given, in an amount proper to pay the costs and damages sustained by the Federal Government, when the injunction or order is issued pursuant to rule 65(c) of the Federal Rules of Civil Procedure after the date of enactment of this Act. No court may consider any factor other than the value of the costs and damages sustained when making its determination of the proper value of such security, and that determination shall be appealable upon issuance of the preliminary injunction or temporary restraining order under an abuse of discretion standard.

This is a vast improvement. The essential point of continuity is that the new bill retains the focus on injunction bonds. But the Senate provision is far more effective at increasing the use of injunction bonds. That is so because of six changes:

1. A shift from regulating all federal court injunctions to regulating only preliminary injunctions and temporary restraining orders. This makes obvious sense, because injunction bonds are simply not used for final injunctions.¹

2. A shift from regulating injunctions against all parties to regulating only injunctions against the federal government. The clearest pattern of non-compliance with Federal Rule of Civil Procedure 65(c)’s injunction bond requirement is in suits against government defendants (see Preliminary Injunction Realism), so this narrower focus makes sense.

3. A shift from restricting enforcement of injunctions to requiring conditions for the grant of injunctions. The former approach (in the House bill) was a likely unconstitutional intrusion on core judicial power, while the latter approach (in the Senate bill) is more squarely within the power of Congress to regulate the jurisdiction of the federal courts.

4. A shift from retroactive immunization to a prospective rule. This shift, too, avoids a constitutional problem and is more consistent with the rule of law.

5. A shift from a rule that is easily evadable—an on-off effect for the provision based on whether there is any injunction bond, which made the House version easily evadable with a $1 bond—to one that is meant to scale the required bond amount with the cost of the interlocutory injunction to the government.

6. An explicit carveout for bankruptcy.

All of these are improvements. The Senate version is dramatically more effective and it removes the constitutional problems with the House version. In effect, this provision requires federal courts to actually follow Rule 65(c) in suits against the federal government.

So where are the points of vulnerability? One is that it remains difficult to come up with bond amounts when the suit (and the requested PI) are about harms that are not easy to calculate. That difficulty is simply intrinsic to injunction bonds in these cases. I discuss it some in The Purpose of the Preliminary Injunction, but also suggest the difficulty can be overstated.

Another is that the provision seems to explicitly rule out considering plaintiff indigency in setting the bond amount. I argue for a contrary position in The Purpose of the Preliminary Injunction. I understand the desire to avoid a loophole that could be easily exploited by litigants and courts. Even so, consideration of the resources of the parties—and in particular making sure that lack of money doesn’t become a barrier to access to courts—is deeply rooted in the law of equity, even if that ideal is not always achieved. When Charles Dickens in Bleak House warned that the Court of Chancery “gives to monied might the means abundantly of wearying out the right,” that was a criticism!

Note I say that the provision “seems” to rule out consideration of indigency because the answer may turn on what “proper” means. I think the most natural reading would be that proper is not importing some kind of independent requirement of conformity with all constitutional norms and values, but hey, there are people who think that’s what proper means. (Yes, that’s a dig at Printz—cf. the hendiadys paper.)

Finally, the actual effect of the provision will depend on how much evidence federal courts require the federal government to put on about their likely costs from a wrongly granted injunction, and also about how committed and capable the Department of Justice is in putting on that evidence. If the federal courts demand substantial proof up front, the federal government may not get more than a nominal injunction bond in most cases. And a federal district court’s demand for evidence from the executive branch about its assertions is insulated on appeal by the abuse of discretion standard.

The precise effect therefore depends on the future interpretation and implementation of the bill, as well as the follow through by the executive branch. It could be that the provision will turn out to be limited in its effect. On the other hand, it could provide a strong impetus to lower courts to require injunction bonds. The latter scenario seems more likely in cases in which there are readily calculable costs to the federal government from a preliminary injunction or temporary restraining order.

But what is not in doubt is that the Senate version is a dramatic improvement over the House version—it avoids the serious constitutional problems that were likely to doom the House bill, and it is more effective and harder to evade in requiring meaningful injunction bonds in suits against the federal government. The effect of that change will be to give more weight in the preliminary injunction calculus to the regulatory cost of preliminary injunctions to the federal government—not just in the current administration, but in future administrations, both Republican and Democratic.

Two last notes:

1. The Senate Judiciary Committee’s proposal also includes an annual appropriation of $1.25 million to gather statistics on federal court dockets, including non-party relief and injunction bonds (Section 201).

2. If you want to read more on injunction bonds and why greater use of them makes sense—apart from who the president is, and simply as a matter of ensuring that federal courts are considering the costs on both sides when awarding preliminary injunctions—see The Purpose of the Preliminary Injunction. One persistent misunderstanding is that an injunction bond does meant that the plaintiff will actually pay any money to the government defendant. In fact, the government will recover under the bond only if the government ultimately wins the case (i.e., the preliminary injunction was wrongly granted) and the government can prove that the wrongly granted injunction caused it losses. And even then the recovery is limited by the proof of loss and by the amount of the bond. So what the injunction bond does is (1) permit recovery by the government for losses caused by the preliminary injunction, should the government ultimately win the case; and (2) cap that recovery at the amount of the bond.

1. One could imagine a one-in-ten-thousand exception where a structural injunction requires performance bonds on one or both sides. So it’s not that some kind of security device cannot be used for final injunctions. It’s that it would be extremely rare, and it has nothing to do with Federal Rule of Civil Procedure 65(c).