There was a time when the Fed Chair’s semi-annual testimony before the House or Senate was something of a love fest. Chairman Greenspan would be celebrated by members on the left and right, each side eager to latch onto the growing perception that Greenspan—and the Fed—was the best thing going in U.S. economic policy.

Judging from the hostility that the Fed faces in those same congressional halls, it is safe to say that those days are long past. The Fed is a target of congressional ire, which has turned it into a target for congressional reform.

But there are two ways to reform the Fed. The first way is to change the Fed’s functions. That is, expand its duties here, contract its duties there, and otherwise change the business of central banking. The Dodd-Frank Act was primarily a functional change to the Fed.

The other way is to change the Fed’s structure. These kinds of reforms have less to say about which levers the Fed will pull and more to say who, in fact, will be changing those levers. From the discussions at Senate hearings, it appears that Congress seems poised to make the next round of reforms primarily structural ones.

The most intriguing set of structural proposals come not from outsiders, but from within the Federal Reserve System itself. Richard Fisher, the just-retired president of the Federal Reserve Bank of Dallas, has enjoyed a big send off in the press (here’s a profile in the New York Times, here’s a discussion of some of his recent proposals in Politico). This is well deserved. Fisher was an important figure in the financial crisis and gave voice to a lot of widely held concerns about financial regulation, the Fed’s response to the financial crisis, and much else. Given his colorful metaphors and his outside-the-Beltway status, he has been seen as the outsider’s insider on the Federal Open Market Committee.

In what was nearly his valedictory address as the Dallas Fed President, Fisher made a fairly strong set of structural proposals that would dramatically change the Fed’s governance with his customary charm and accessibility. (The speech is called “Suggestions After a Decade at the Fed: With Reference to Paul Volcker, Roosa Boys, Hogwarts, the Death Star, Ebenezer Scrooge, Mae West, Herb Kelleher, Worms and Camels, Peter Weir, Charles Kindleberger, Pope Francis and Secretariat”; and, if anything, he understates the number of metaphors, anecdotes, folk wisdom, and color in the speech.)

Fisher introduces the proposals with “an ancient Arab saying that one should ‘trust Allah but tie your camel.’” His point is that we need to protect the need for an independent monetary policy without compromising the very real need for democratic accountability. Fisher thinks—rightly, I think—that various proposals to subject the Fed to searching congressional oversight over the day-to-day of monetary policy are bad ideas. But his proposals to “tie the camel,” misunderstand the best defense of central bank independence, and make the Fed more opaque, rather than less. The Senate should reject them, with one minor exception. Let’s take them each in turn.

First, Fisher thinks the Vice Chairmanship of the Federal Open Market Committee—the Fed’s key monetary policy-making committee—should rotate on a two-year basis among the twelve Federal Reserve Banks. Currently, by custom, the New York Fed president is the permanent Vice Chair of the Committee, making him a dominant force on the Committee, second only to the Fed Chair.

I agree with Fisher’s diagnosis of the problem, but not with the solution. New York’s bankers, who exercise an outsized influence on the selection of the New York Fed’s president, should not get this kind of seat at the table. But why should be solicitous of the bankers and businessmen of Kansas City, Dallas, or Cleveland? Fisher, for example, goes on at colorful length about how close he is to the business leaders in his district. I have no doubt that the founder of Southwest Airlines is an able executive and entrepreneur. But why should he get direct access to our nation’s monetary policy? There is no justification in the theoretical or empirical literature on central bank independence that would give this kind of access; there is even less defense from the perspective of democratic governance and accountability.

A much better solution would be to end the opaque and multi-level governance structure that creates two separate Vice Chairs at the Fed, one for the Board of Governors, one for the FOMC. Currently, the Vice Chair of the Fed’s Board of Governors is Stanley Fischer. Fischer (note the “c”; no relation to the outgoing Dallas Fed President) is one of the most eminent scholars of macroeconomics of the last forty years. His students have been in their own right huge figures in macroeconomics (including Ben Bernanke, Mario Draghi, and Greg Mankiw, among others). And he is the former Governor of the Bank of Israel, where he won high marks for leading the Bank and the Israeli economy through the thick of the financial crisis.

As impressive as these qualifications are, they are not Stanley Fischer’s most important qualification. His most impressive credential is that the U.S. President nominated him and, after a not uncontested nomination process, the U.S. Senate confirmed him. Nothing like the same can be said of any of the Reserve Bank presidents. Rather than have the FOMC’s Vice Chair be a sinecure for those who represent their local business districts in twelve cities selected more for their political benefits to the Democrats in 1913 than for their financial prominence, the Vice Chair of the Fed should be the one the people’s representatives selected to be the Vice Chair of the Fed.

Second, Richard Fisher thinks that the regulation of systemically important financial institutions (SIFIs) should be regulated by a Reserve Bank outside its district. So the Dallas Fed would regulate SIFIs in New York, the New York Fed would regulate SIFIs in Kansas City, etc. The reason for the proposal is that there is a fear that the Reserve Banks’ own proximity to the regulated banks compromise their ability to regulate. They become “captured,” in the regulatory parlance.

Again, I see the point of the problem, but the solution strikes me as, once again, confusing and wrong. The point of governance is to allow some kind of distance between the ultimate reservoir of authority—in a democracy, that’s all of us—and those who execute the policy. Governance is about the rules of the delegation. Making the Fed even more organizationally and arbitrarily complex means that there will be even less ability for the people and our representatives to know who is responsible for what inside the Fed. If there is a huge regulatory failure at a New York bank, will the public know that the Dallas examiners were the ones partly to blame?

Instead of scrambling the eggs of the Fed’s regulatory apparatus, I would favor essentially removing the Reserve Banks from the business of SIFI supervision altogether (this appears to be the practice, post-2010). There will, of course, be capture concerns at the Washington-based Board of Governors, as there always will be in any part of the bureaucracy. But the difference is that we and our representatives can participate in the appointment process at the Board of Governors; we can’t for the Reserve Banks.

Fisher’s proposals go further in this wrong direction from here. He next proposes giving the Reserve Banks’ parity with the Board of Governors on the FOMC. Currently, the U.S. President-appointed, Senate-confirmed Board of Governors enjoys a 7-5 majority. But this is, again, a compromise that reflects politics long passed. And retaining the current Reserve Banks on the FOMC is not something I think anyone, if they were designing the Fed from scratch, would pursue. Two Reserve Banks in Missouri? Only one west of Dallas? To say, as many have, that the Reserve Banks reflect regional, democratic legitimacy makes no sense. That’s a very curious theory of regions and a very curious theory of democracy.

Instead—although I recognize this is a political impossibility—the better solution is to remove the Reserve Banks from the FOMC entirely. There’s a constitutional problem with their ongoing participation, as I have argued elsewhere. But more than that, it would be refreshing to flush from the system the vestige of a political compromise long passed that few understand.

Fisher’s last proposal is the only one with which I agree. He wants to have the Fed Chair hold a press conference after every FOMC meeting. I think this is a very good idea. The reason is not simply to explain what the Fed just decided, although this is very useful too, but to present the Fed to the press for questions that reflect the economic and political zeitgeist. We get some of this through the Chair’s regular testimony before both Houses of Congress, but congressional hearings aren’t always the most efficient way of disclosing information. Press conferences give much to Fed accountability and take little from Fed independence.

Fisher and I agree that Fed independence is an important concept, that we want to keep the regulation of the currency outside of the day-to-day of political pressures. It is often said that independence and accountability are reciprocals, and that we must optimize those tradeoffs. But this isn’t exactly true. There are ways to increase accountability of the Fed generally without compromising the kind of independence that is useful. And that way is through simplifying the Fed’s governance by making it clear who wields the Fed’s power, how, and to what end. President Fisher’s recommendations scramble that governance structure even further. His proposals should be rejected.