By timing the disclosure of the results of its vaccine trial, Pfizer could have influenced the 2020 presidential election. This is worrisome regardless of whether Pfizer used that power or not.


Most people consider it self-evident that corporate campaign contributions impact policy decisions. In fact, when he was running for president in 2016, then-candidate Donald Trump openly bragged about his ability to influence politicians: “When you give, they do whatever the hell you want them to do.” Ironically, what is self-evident to most people has failed to gather much academic support in the political science literature. A forthcoming article in the Journal of Politics finds no significant evidence that firms benefit when the candidate they backed financially wins a highly-contested election. This is just the latest in a long list of academic articles with similar findings. Thus, should we conclude that the concern regarding firms’ excessive political power is vastly exaggerated? 

No. On the contrary, we should conclude that political scientists (like economists) prefer to look under the lamppost where the light (i.e., the data) is. To resolve the apparent paradox, political scientist Paul Pierson advocates a more case-based approach. Only by looking at the full context can we expose where corporate power manifests itself. The timing of Pfizer’s vaccine announcement is an excellent application of this idea. President Trump alleged that the pharmaceutical company deliberately waited until after the presidential election was called to announce the promising results in an effort to hurt his re-election chances. Much of the press, including the prestigious scientific journal Science, dismissed the claim as baseless. Nevertheless, the evidence seems to suggest otherwise. We will argue that Pfizer had both the opportunity and the motive to influence the 2020 presidential elections. 

The announcement was scheduled to come out before the election

Pfizer vaccine trial began at the end of July 2020. Following industry best practices, Pfizer delegated the responsibility to examine the clinical data to a data monitoring committee (DMC) composed of independent scientists. In the original protocol, Pfizer would ask the DMC to evaluate the vaccine’s effectiveness after 32, 62, and 92 people in the sample became infected. The date at which these milestones would be reached depended entirely upon the speed of diffusion of the disease in the areas where the subjects lived (the United States and Europe).

As recent as mid-October, the expectation was that the first milestone would be reached by the end of October. Pfizer’s CEO Albert Bourla stated this in an interview on September 6 and then reinforced it in an open letter published on October 16. This view is also consistent with the fact that in late August, the CEO revised his stock sale plan, pre-committing to selling a significant amount of shares on November 9. These plans allow insiders to unload their stock holdings without generating suspicion of insider trading. Precisely for this reason, the trading is either distributed over time (as in the case of Moderna’s CEO) or, if large amounts are sold at specific dates, the best practice is to keep those dates far away from major corporate announcements. Since the plan’s revision was done after the start of the Covid-19 vaccine’s phase 3 trial and following the protocol’s timeline, it is reasonable to assume that Pfizer anticipated disclosing the trial’s interim results at least a week before November 9. 

Pfizer used their discretion to delay the announcement after the election but before November 9 

Why, then, was there no announcement until November 9? It is certainly not because the spread of the disease was below expectations. In fact, Kathrin Jansen, head of vaccine R&D at Pfizer, told Science Insider that with the pandemic spiraling out of control, “we probably could get cases much faster than what we had anticipated.

The reason is that Pfizer preferred to forgo analyzing the data at the original 32 cases and petitioned the FDA to perform the first analysis at 62 cases (almost certainly after the elections). The reason for this change is not explained nor justified in Pfizer’s press release, which simply states, “After discussion with the FDA, the companies recently elected to drop the 32-case interim analysis.” While this might be an excellent business decision (more data can show a higher efficacy of the vaccine), this decision was not imposed by “science” nor mandated by regulation. On the contrary, this was clearly a strategic decision by the firm and its CEO. 

Pending FDA approval for this protocol change, the company went to the extreme act of storing the nasal swabs taken from participants rather than testing them directly. So much so that by the time the DMC received the data, the total infected were 94, surpassing the third milestone.

“The announcement came just hours before the CEO’s shares were scheduled to be sold, ensuring him more than half a million in extra profits.”

The data was presented to the DMC on Sunday, November 8, allowing it to analyze the data by the end of the weekend and announce the positive findings first-thing in the morning of November 9. The announcement came just hours before the CEO’s shares were scheduled to be sold, ensuring him more than half a million in extra profits. 

Pfizer likely to gain from Trump’s loss

Pfizer certainly had the opportunity to delay the announcement of the vaccine strategically. But what about the motive? In September, Trump announced an executive order requiring Medicare payment for medicine to be capped as the lowest price charged by pharmaceutical companies to other developed countries (the so-called most favorite nation clause (MNF)). Contrary to expectations, this order was not limited just to Medicare part B, but also included Medicare Part D, potentially costing billions of dollars a year to the pharmaceutical industry. As John McManus, president of the lobbying firm The McManus Group—which is very active in the pharmaceutical space—stated on October 12, “a Democratic sweep in the election is likely to result in a Biden administration withdrawing the MFN rule.”

Conclusions

Pfizer had both the motive and the opportunity to influence the presidential election. Did they do it? We do not know, and it is irrelevant for the point we want to make. If firms such as Pfizer have the power to swing elections, whether through campaign donations or the strategic release of information for their own advantage, that in itself is worrisome regardless of whether that power is used or not. As economists, we know all too well the power of out-of-equilibrium threats. The mere threat to wield this influence can have significant policy consequences. Future presidents beware: you challenge the pharmaceutical industry at your own risk!