*This is the fifth post in a series on Andrew Rudalevige’s new book, By Executive Order: Bureaucratic Management and the Limits of Presidential Power. For other posts in the series, click here.

It was clear early in the development of the unilateral powers literature that focusing on executive order issuance posed a selection bias problem. By definition, we observe executive orders when they have been issued. Because not every proposal or idea for an executive order turns into an issued order, focusing on the ones that made it into the Federal Register meant that we would not observe what happened between a proposal and a decision not to issue an actual order. This is hardly a mystery, and previous analysis of executive orders included case studies of orders that were never issued or ultimately got revoked (Mayer 2001; Thrower 2017), or incorporated into formal models measures of discretion that might make a president reluctant to issue an order (Howell 2003).

But Rudalevige advances the state of knowledge (considerably). By situating these preliminaries at the center of the inquiry, Rudalevige’s By Executive Order makes an enormous contribution to scholarship. His comprehensive analysis of what occurs before an executive order is issued, and in particular the circumstances in which executive orders wind up not being issued adds both depth and breadth to our understanding of unilateral action. It also connects to other literatures on regulatory review, the importance of staffing, and the nature of bureaucratic governance.

I was struck by how strongly the book is linked to other key observations in presidency studies that are not part of the unilateral powers literature.

“The president is not the presidency” (Jones 2005). With those six words, Charles O. Jones captured something about the presidency that is at once a truism and something deep and complex about the office (or person, or institution, or organization, or staff, or any one of a number of different possibilities, which gives some idea of the depth of the problem). A president may at times have the authority on paper to order others in government to “do this! Do that!” But even when that authority is unambiguous – and over the last 20 years the unilateral powers literature has established that it often is – carrying out anything other than the most trivial order requires many others besides the president to do the actual implementation.  Even when the president acts alone, the president doesn’t really act alone. As Neustadt noted 60 years ago in Presidential Power, there really isn’t any such thing as a “self-executing” order.

No one who studies the presidency will be surprised that staff matters. But Rudalevige’s analysis of how presidents must aggressively manage the transaction costs that result from centralization confirms Samuel Kernell’s observation more than 30 years ago that the same structures that help the president deal with management and monitoring problems across the executive branch force the president to deal with the management and monitoring problems across the White House (1989, 236-7). Internalizing transaction costs by central coordination does not reduce those costs to zero, and sometimes increases them.

And we have seen, with Donald Trump, what happens when a president does not care about process; he shows what happens when the presidency is, in fact, reduced to the president. The consequences for getting his preferences implemented – to say nothing about good government principles – are obvious: a White House paralyzed by infighting, astonishingly sloppy staff work, executive orders issued and reissued, nobody quite sure whether a tweet constitutes a formal policy statement or direct order, and a catastrophically mismanaged response to COVID-19 that likely cost him reelection

A presidential candidate who says “if I am elected, I will issue an executive order to make this happen” will get more cheers than one who says “if elected, I will initiate agency processes to investigate options for acting on this issue, taking care to balance concerns by staff that the effects of any order not conflict unduly with other goals, making sure that lawyers in the White House, Office of Management and Budget, Office of Legal Counsel in the Department of Justice, and the affected agencies all have an opportunity to suggest language that is most likely to produce the desired effects, incorporating the concerns of affected public constituencies, and following up with appropriate attention and resources to ensure that any order I issue is actually implemented.”

But presidents who understand the difference are more likely to get what they want. As will presidents who read this book.

Kenneth R. Mayer is Professor of Political Science at the University of Wisconsin – Madison.