Can Apple and Google Rely on Trump's Promise Not to Sue for them for Hundreds of Billions of Dollars?
On Friday of last week, I argued that President Trump's Executive Order (EO) purporting to suspend implementation of PAFACA, the law banning TikTok so long as it is owned by ByteDance or another Chinese company, was not within the authority Trump claimed for it in the EO itself. However, I explained that Trump and whoever drafted the EO might have done better had they relied on prosecutorial discretion to forbear bringing enforcement actions based on conduct during the 75-day period. My bottom-line conclusion, though, was that the promise of enforcement forbearance was not enforceable, which explains why Apple and Google have not been assuaged by the promise and thus have not restored TikTok to their app stores.
In response to my assertion that the promise is not enforceable, I received skeptical emails from two readers whose legal acumen I greatly respect. Why wouldn't it be an enforceable promise?, they asked. After all, the government enters contracts; it enters plea bargains; etc. Why isn't this promise enforceable? And wouldn't it violate due process for the government to announce that it won't enforce a law and then turn around and enforce the law against parties who violated it in reliance on the promise?
I answer these questions today. Although I continue to believe my analysis to be correct, I don't think it's a slam-dunk. Nonetheless, I think there is at least very serious doubt about the enforceability of the promise of forbearance in the EO. If there were not, then Apple and Google surely would have restored TikTok to their app stores. After all, keeping TikTok off the app stores while the EO is in effect risks annoying Trump, which, given his track record of petty vindictiveness, could result in his punishing them on unrelated matters. The fact that they are willing to risk alienating Trump in that way supports the inference that they believe they face a serious risk of crippling liability if they violate PAFACA in reliance on the EO. And it should go without saying that Apple and Google have access to high-quality legal advice.
Turning to the merits, why isn't the promise made by the EO an enforceable contract? Law students learn in their first semester that a contract generally requires an exchange--what the law calls "consideration." If Al promises Barbara his bicycle, that's an unenforceable promise of a gift. If Al promises Barbara his bicycle in exchange for $200, that's a contract. Courts have long recognized that, even in the absence of consideration, a promise may be enforceable if the promisee detrimentally relies on the promise. Not all detrimental reliance counts, but the doctrine of promissory estoppel is well established in contract law--including the federal common law that applies to a small category of cases. (Contract law is generally state law. There is no general federal common law, but in certain settings of federal interest or power, federal courts apply a specific federal common law.)
But here's the thing. Promissory estoppel is harder to establish where the promisor is the government rather than a private party.
As the Eleventh Circuit put the point in a 1988 case, "for estoppel to apply against the Government (1) the traditional private law elements of estoppel must have been present; [and] (2) the Government must have been acting in its private or proprietary capacity as opposed to its public or sovereign capacity." However, in enforcing (or rather, not enforcing PAFACA), the government is clearly acting in its public or sovereign capacity, not like a market actor making a private bargain. That strongly indicates that promissory estoppel would not be available to Apple or Google. And because there's no conventional consideration, that means the forbearance assurance in the EO is unenforceable.
What about plea bargains as a model? The Supreme Court has held that a plea bargain is an enforceable contract, although the rules of contract don't translate exactly. As the Court said in Ricketts v. Adamson, "Unlike some commercial contracts, plea agreements must be construed in light of the rights and obligations created by the Constitution." But the Court has never suggested that there's any reason to think that an agreement that wouldn't be enforceable in the commercial setting is enforceable as a plea bargain.
And that brings us back to the earlier point. A plea bargain does include mutual consideration. The defendant agrees to plead guilty in exchange for the government's agreement to pursue a lesser charge or lesser sentence than it might be able to achieve in a contested trial. Plea bargains do not involve promissory estoppel--and even if they did, as we have seen, a crucial element of promissory estoppel is not present in the TikTok EO, which exercises the government's sovereign power, not private or proprietary power.
Although not suggested by my readers, we might also consider a broader equitable estoppel. Even apart from contract law, there is undoubtedly something inequitable about the government telling a firm it won't be liable for some action and then holding the firm liable when the firm takes action in reliance on that assurance. Equitable estoppel strikes me as potentially applying, but the courts are quite reluctant to apply it against the government.
Heckler v. Community Health Services of Crawford County, Inc. (CHSCC) is a good example. There, the Supreme Court rejected equitable estoppel where the respondent expended resources in reliance on the assurance of a third-party intermediary that the expenses were reimbursable under Medicare. The assurance turned out to be false. The Court rejected the estoppel claim anyway.
Admittedly, the argument for justifiable reliance on an assurance of no liability from the president and attorney general is stronger than for reliance on a third party, even if that third party is an agent of the government, as the Court treated the third party in CHSCC. Nevertheless, the Court's opinion in CHSCC contains some pretty sweeping language that sets a very high bar for equitable estoppel against the government:
When the Government is unable to enforce the law because the conduct of its agents has given rise to an estoppel, the interest of the citizenry as a whole in obedience to the rule of law is undermined. It is for this reason that it is well settled that the Government may not be estopped on the same terms as any other litigant.
Indeed, near the very end of the CHSCC opinion, one finds a suggestion that estoppel might never be available against the government. The penultimate sentence of the opinion begins by "assuming estoppel can ever be appropriately applied against the Government . . . ." If the Court needs to assume estoppel can ever be applied against the government, that implies that it was leaving open the possibility that it can't ever be so applied. The possibility of equitable estoppel does not provide sufficient assurance that Apple and Google wouldn't face billions of dollars in fine.
I should say that I share the sense that it does seem unfair for the government to lure parties into violating the law and then (criminally or civilly) prosecute them. But I'll add three points by way of conclusion.
(1) The evaluation of unfairness is channeled through the doctrines I've discussed above. Perhaps promissory estoppel or equitable estoppel should be more freely available against the government, but until the law changes to make it so, a general sense of fairness won't provide legal assurances.
(2) The unfairness of the scenario I'm envisioning is not unique to the executive branch. In Edgar v. MITE Corp., Justice Stevens argued in a partial concurrence that a federal district judge lacks the power to grant pendente lite relief.
I can illustrate with a hypothetical example. Suppose Penelope wants to hold a march in violation of a local ordinance that, she believes, violates the First Amendment. She sues the District Attorney in federal court and obtains an injunction against the ordinance's enforcement. While the injunction is in place, she holds the march. The D.A. appeals and the appeals court reverses, holding that the ordinance didn't violate the First Amendment after all. According to Justice Stevens, the D.A. can now prosecute Penelope for the march that occurred even while the ordinance was enjoined. Justice Marshall disagreed, and the majority didn't address the issue, but it deemed the matter serious--suggesting that there may indeed be no district court authority to grant protection pendente lite.
Where executive action is at stake, there isn't an appeal to a higher authority, but we can analogize Trump changing his mind or a subsequent administration bringing an enforcement action to the appeals court reversal in the judicial case.
(3) Finally, despite the unfairness to Apple and Google in the scenario in which they rely on the promise of non-enforcement to their detriment and then end up being sued for hundreds of billions of dollars in civil fines, it is not at all clear to me that the authority to make binding promises of non-enforcement of the law is a power a president has or should have.
Suppose that on January 6, 2021, Trump had explicitly promised pardons to anyone who was charged with any crimes arising out of storming the Capitol to stop the counting of Electoral votes. Suppose that he had then not issued such pardons before leaving office. Surely, no one then charged could say that they relied on the promise of a pardon and therefore they cannot be prosecuted. If that's right, then the same should be true if, instead of a promise of a pardon, Trump had issued a promise of non-prosecution, even if he did so in letters given to individual insurrectionists. But that is essentially what Trump is trying to do for Apple and Google, albeit promising non-enforcement of a civil penalty rather than a criminal penalty.
Thus, extant law casting very serious doubt on assurance given by the TikTok EO aligns with sensible limits on a president's power to authorize lawbreaking.