Is this on? It's on now. Okay. All right. Yeah. Thank you. That was unnecessarily generous, but I'll take it anyhow. So, and I'm impressed that this many people are here to hear a talk about bureaucracy, because generally speaking, if you utter that word, everybody runs in the other direction and tries to, you know, looks at their phone and pays attention to, you know, almost anything other than this topic. But it is one that I actually have done a lot of work on and care a lot about. So this is going to be a different kind of talk from the usual weekly seminar. It is not a report about a research project that I am engaged in. It's a little bit different. I've been writing a little autobiography over the last year, or kind of autobiographical memoir. And one of the things I've been trying to do is... so I've written 10 single-authored books, and there's actually a lot of connections between these books that are not obvious if you haven't been really paying attention. And so what I'm going to do is actually follow a trail of crumbs, because there's a bunch of ideas in my different books that are all connected to each other. And I'm going to try to explain to you what those connections are. And it's one that leads me from a kind of theory of hierarchy all the way to DOGE and what's going on in Washington, D.C. now. Okay. So that's the plan. By the way, for the students, I'm happy to share these slides. There are a lot of citations here. So if you actually want to follow the actual articles that I'm building on, you know, you'll have the full citations. All right.
So, hierarchy is a necessary fact of human organization, and there's a lot of theory about hierarchy. I would say that the dominant theory really comes from Ronald Coase. He wrote an article in 1937 called 'The Theory of the Firm.' Coase is an interesting fellow. He actually never wrote any books. He won the 1991 Nobel Prize in Economics, although he's actually a sociologist, and he's famous for really two articles. So the first one was this article, 'The Theory of the Firm,' in 1937. Coase raised the following question: If you believe in a market economy, everybody should just be dealing in a horizontal way with other market participants. Why is it that you've got these big hierarchical corporations like the Ford Motor Company or General Motors? At that time, you know, those motor companies actually, they mined iron ore, produced steel, they made cars, they marketed cars, you know, they did everything from soup to nuts. And he said, 'Why couldn't you do this in a decentralized way, basically outsourcing every stage of the automaking process to different firms? Why do you have these big hierarchical organizations?' And his answer was very simple: It's transaction costs. That the cost of coordinating all of that kind of outsourcing would be too expensive, and it's much easier to do it just through a single authoritative hierarchy. And there's a whole literature in economics about the theory of organization that begins with Coase's theory of the firm. Oliver Williamson is probably the most prominent person writing in that vein. Now, there was a libertarian critique of this. A lot of people don't like the idea of hierarchy, and especially, you know, in the 1990s, early 2000s, when we're in this big neoliberal, you know, market-oriented period in economic thought, there's another article that Coase wrote. I believe that this was the single most cited economics article in history, 'The Problem of Social Cost,' where he begins with this classic problem of farmers and ranchers, where the ranchers let their cattle wander over the farmer's property, and it, you know, they do a lot of damage. And the idea was that they had to have liability, you know, to protect their property rights. And Coase basically said, well, if you could imagine a world of zero transaction costs, a government wouldn't be necessary because the farmers and the ranchers could just negotiate with each other, and they would come to a just, you know, division of property rights, and everything would be fine. So again, transaction costs played a big role in this. A lot of economists and libertarians love this idea that you actually didn't need a hierarchical organization to, you know, assign rights in this kind of a social dispute. You could simply work it out in a kind of horizontal coordination, right? So these two articles were quite related to each other. There's an article by an economist at MIT named Thomas Malone that was written in around this period when the internet was being privatized, and his argument was, well, if the problem is transaction costs that's keeping all this horizontal coordination from happening, the internet is great because you can negotiate things, you know, you pass information horizontally. And the old distinction that Coase had made between markets and hierarchies, he said, well, there's actually a third layer in between there, which are networks. And those are all different ways of settling kind of social organizational problems. So the 1990s saw all these people trying to figure out, could we get to social organization without a hierarchy? The idea of a complex adaptive system became very popular. You had a whole organization, the Santa Fe Institute, that was devoted to studying it. It's really a fascinating idea because, you know, they wanted to focus on the idea of emergent order. A flock of birds does not have a bird president or a bird king that tells the birds where to fly. They simply have a very simple algorithm, and it leads to the emergence of order, you know, when the flock flies around looking for food, something like that. And so the idea was maybe you could use complex adaptive systems to actually produce social order without having to have a central authority, right? So these were all ideas that were very fashionable in the 1990s, the early 2000s. It was all stupid. Honestly, if you've ever tried to do anything through a consensus, make, you know, a body that requires consensus like NATO or the European Union, you can't do it, right? It's too hard to negotiate actually common positions this way. Whoops, sorry, happened. But the earlier thinking about the firm led to, you know, what we now call principal-agent theory, which I think remains the dominant way that economists think about hierarchical organization. So you couldn't get away with hierarchy, but I think that there is a lot of interesting writing that, you know, in a way questioned principal-agent and said that that might actually not be the best way to understand how organizations work. You know, principal-agent said the principal makes a decision, makes mandates. The agents have to do what the principal wants. Corruption and organizational dysfunction occur when the agents have their own agenda, and they, you know, they're lining their own pockets instead of the public treasury, so on and so forth. And therefore, you know, that's where you get things like stock options, where you align the incentives of the agent with that of the principal, so on and so forth. Now, I think that actually there are some pretty serious critiques of principal-agent made by a number of people. One of them goes all the way back to Herbert Simon's famous book from the 1950s on administrative behavior, where he said that in many hierarchies, authority actually doesn't go from the principal to the agent. It actually goes from the agent to the principal because the agent actually knows more than the principal. And I think any of you that have worked in a big organization knows that that's true. The worker bees at the bottom of the organization that are actually doing stuff actually know what's going on on the ground. And the boss, you know, that's sitting in corporate headquarters has no idea. And they require, you know, authority and advice from the agents in order to be able to run the organization. If you've ever seen 'Yes Minister,' this famous BBC comedy from the 1980s, where the bureaucrat Humphrey is really controlling the minister, and then the minister becomes prime minister, but he's still a complete puppet of his agent, right? I think that that's basically what they were getting at. John Dilulio, a great expert on public administration, also pointed out that actually principals don't control agents. Oftentimes the agents control themselves as a matter of norm. You know, this article that he wrote on principal-agent actually saw that in many hierarchies, really the authority does start at the bottom of the hierarchy. So there's a lot of questioning about whether principal-agent was actually the best way to understand the workings of a bureaucracy.
Now, the other important concept here, which actually I've decided is the center of any theorizing about organizations, is the question of delegation. And to see the importance of delegation, I would point to this 1945 article by Friedrich Hayek, which I think is actually one of the most important things he ever wrote, 'On the Use of Knowledge in Society.' So what Hayek argued, you know, he was having this big fight at the time with Joseph Schumpeter as to whether a market economy or a centralized planning was more efficient. And after the 1930s, it wasn't obvious. You know, now we think that markets, of course, they're better than centralized planning, but at that time it wasn't obvious given the experience of the Great Depression. And Hayek said markets are going to win out because, you know, 99% of all the information in an economy is local in nature, and the people that actually know what's happening are at the bottom. You know, they're the decentralized agents that are buying and selling. It's not some planner sitting in a central ministry, right? In the Soviet Union, they had this organization called Goskomtsen, the State Committee on Prices, that set prices for the entire Soviet economy. And, you know, Hayek was right about this. He said that's impossible. A bureaucrat can't possibly set accurate prices reflecting relative scarcities. Those have to evolve. So this was an argument in favor of market economics. But it's also, it leads to a big insight about organizations, which is that organizations have to delegate. They have to delegate because the vast majority of knowledge that an organization deals with and information is local in nature. You can't react quickly if you have to send a decision up a hierarchy and wait for a vice president to make a decision. If you can actually empower people at a local level to act on their immediate information, that organization is going to do better. But the problem with all management, in my view, is the fact that once you delegate, you lose control, and every organization has to figure out how do we keep control over the behavior of the organization as a whole and yet make use of local knowledge, empower level agents within the operation to act. So there's several mechanisms. This again ties together some of the stuff that I've written about. So you can have formal organization, right? You can have a table of organization, you can have job contracts and so forth. And there's also informal ones where if you delegate authority, you are delegating it not on the basis of a job, you know, a wage contract, but you're actually trusting the individuals because you have a good sense that they will keep the interests of the organization in mind. And so things like social capital, education, organizational culture are important ways in which organizations control agents. It doesn't have to be through, you know, monitoring and accountability. I like these words, transparency and accountability. How many times have you heard that as positive, you know, ideas? Basically, what that means is we're going to monitor you and we're going to punish you if you do the wrong thing. That's what transparency and accountability mean. That's a formal mechanism. That is one way of controlling people in a hierarchy. But it's actually much better if your people at the bottom of the organization are trustworthy and you can safely delegate authority to them. Now, that's why I wrote a book called 'Trust.' I think that trust actually is, you know, pretty critical to the proper functioning of a hierarchical organization because if you have to work simply through authority, formal authority structures, it gets very, very rigid, you know, very quickly. Now, so I'm now, I'm going to tell you two stories which to me illustrate why delegation is really important. The first one really had to do with lean manufacturing or just-in-time, the Japanese kanban system. So this was a system invented by Toyota in the 1960s to make their manufacturing process more efficient. What they did was, I mean, they reduced inventories. They only kept inventories that were necessary to, you know, for a day or two instead of piling them up like many American car companies did. But the more important thing was that they empowered the assembly line workers to basically control quality. And so if they got a part that didn't fit, you know, like you're fitting a door panel onto a chassis and you realize that the part you're getting from the subcontractor isn't right, each worker had a cord at their workstation, and if they pulled the cord, the entire production line would shut down. And what this does is creates an enormous incentive to get the quality right. In an American car company, you know, you'd be much more sloppy about this. And then you'd have a rework area at the end of the assembly line, and they'd try to catch the mistakes after they had been made. What Toyota figured out is that if you empower workers, if you trust workers to look for these mistakes themselves, then they will catch them before they are embedded in the final product. And that's one of the reasons that Toyota developed this incredible reputation for efficiency and also for high quality. Now, if you think about that, it requires a huge amount of social trust, right? If you tried to do this in the Britain of the 1950s, when you've got these poisonous labor, or Italy, or, you know, any number of European countries where you've got these really poisonous labor-management disputes, the workers would be pulling the cord all the time just to screw the managers, you know, say, 'Hell with you, you know, we're going to stop work.' So the Japanese system only worked because you had a high degree of trust between the managers and the worker bees in the factory, and you could trust the workers to work in teams to figure out how to fix production problems on their own without having to be told by a boss. Now, the other thing that really underlined the importance of delegation to me, this was the last study that I wrote with my friend Abe Shulsky when I was at RAND, 'Virtual Corporation and Army Organization.' So in the late 1990s, everybody was talking about Silicon Valley and flat organizations and horizontal coordination, and we did a study for TRADOC, the Training and Doctrine Command in the US Army, whether you could apply these principles of flat organization to the classic hierarchical structure of the US Army. So we went all over the Army. We talked to a lot of people. I actually interviewed Dave Petraeus after he jumped out of an airplane in an airborne exercise. He was back then just a battalion commander, a colonel. But in any event, it was a really fascinating journey of discovery because what we found out is the Army did not need to be taught about flat organization, and that actually they had a lot to teach the rest of the society. So what had happened after Vietnam was that the Army recognized a lot of failures in its command structure. All the fragging of, you know, junior officers by enlisted, very poor discipline. So the Army sat down, they went through a review of their entire doctrine. They adopted what was known in the Wehrmacht as the Auftragstaktik, which is basically the German practice of delegating the maximum amount of authority to the lowest command level possible. And if you think about it in a military situation, you know, you're trying to assault a village. You got a platoon leader that's right in front of the house that you're trying to take over. Who's going to know more about that? Right? It's all this local knowledge stuff. Who's going to know more? Is it the second lieutenant in front of the house, or is it the general, you know, a hundred clicks back at headquarters? Obviously, it's a platoon leader. And so the Army said, okay, we are going to adopt a system that goes under different titles like mission orders or commander intent. The commander's job is only to focus on big strategic or operational issues, and every other decision gets to be made by the lowest possible command level in the Army. That's what makes for a good military organization. That's what gave the IDF a big advantage over much more centralized Arab armies in the various wars that they fought. If you look at Ukraine versus Russia, Ukrainians actually adopted a lot of mission orders kind of doctrine, and the Russian army has remained highly centralized in that Soviet fashion, and therefore they were not permitted to actually take the initiative when they saw an opening. Like a lot of examples of this, like the Third Armored Division when it got up to Baghdad, they were expecting, the planners said it's going to take a week of fighting in order to take Baghdad, and the local commander said, hey, there's nobody in front of us, and he actually didn't wait for orders. He simply took the city, you know, and he did it in one afternoon because he was empowered, you know, to have the authority to make that kind of an important command decision without referring it to, you know, to headquarters, right? So all of this, you know, these two examples really cemented in my mind this idea that you really need to delegate authority, but the process of delegation is extremely complicated because you can also delegate too much authority. There are many examples of, like, I remember at the time with Barings Bank, this big financial firm, you know, collapsed because they basically gave a low-level trader the authority to bet the firm, which he did, and he lost, and the whole thing went bankrupt, right? So you don't want to just willy-nilly delegate, you know, that kind of authority. And I think that all of management, basically, in any private or public organization, depends on getting that delegation formula right so that you can keep control, you know, protect the organization as a whole, but make use of local knowledge and, you know, all the things that make, you know, delegation important. All right.
So, how do you, how much to delegate? There's actually not a lot of theory about this. And I would say that political scientists have in general been a little bit more positive about delegation than either administrative lawyers or economists. So this is just a list of famous political, well, not all of them are famous, but Peter Evans, you know, wrote about embedded autonomy, about how the Japanese developmental state, you know, actually delegated authority to economic technocrats. Dan Carpenter wrote about the delegation of authority in 19th century, you know, the 19th century growing administrative state. One of our former postdocs, Rio Matsuzaki, basically wrote a book about how the Japanese colonial administration in Taiwan had been delegated a lot of authority. And so the basic decisions about how to govern Taiwan were not made in Tokyo. They were made by the local guy on the spot. Dan Honig wrote a book recently about how development bureaucracies are much better when they're actually controlled by people in the field rather than by people in headquarters. Right? So there's a substantial political science literature that thinks that more bureaucratic autonomy will make organizations work better than, whoops, sorry, than others. So this led me to puzzle about this because in the development realm, there was a lot of feeling that you shouldn't delegate stuff. So Bob Klitgaard, who's like, maybe still is like the leading authority on corruption, had this formula that's cited by many people, World Bank really led this, said corruption is equal to monopoly plus discretion minus accountability, meaning the more monopoly and the more discretion you gave a bureaucratic agent, the more corruption there would be, and therefore you wanted to take away as much discretion as possible. This led, you know, for example, to the Indonesians outsourcing their highly corrupt corruption agency to a Swiss company in Singapore so that actually the agents couldn't have any discretionary authority. And, you know, corruption agencies are, I'm sorry, customs agencies are famous for their corruption because, you know, you're making all these judgments, this container can go, this container can go, oh no, I'm going to inspect this container, right? And so it was very easy to bribe people in that situation, and it was so bad in Indonesia that they simply took away all discretion whatsoever. Right? And so the World Bank was pushing all their clients to eliminate discretion, whereas in my reading, discretion was necessary. It was a good thing. You know, you wanted people to have the authority to make local decisions. So how do you square this circle? So this led to an article that I published in Governance. So I won a prize last year. It was a lifetime achievement prize for work in public administration by the American Society for Public Administration. And I think it was largely for this article, which my political science friends may not have read, but it was a theory about what's the appropriate level of discretion to grant a bureaucratic agent. And I postulated that there is a kind of inverted U where too much discretion is a bad thing and no discretion is a bad thing, and there's a kind of optimal level of discretion, which is the, you know, the top of that inverted U. So the question is, where do you put the top of that? And my theory is summarized in this chart. It all depends on the level of capacity, of state capacity. If you have a very low capacity bureaucracy, you want to take away discretion, and if you have a very high capacity bureaucracy, you want to provide much more discretion. So I think you could illustrate this in airline security. So when after September 11th, the United States created the TSA, they deliberately didn't want to hire expensive employees. And so everybody in TSA basically just has a high school equivalency degree, and they have no discretion at all. Right? They can't say, 'Oh, this person looks suspicious. We're going to pull you out.' Everybody has to take off their shoes. So they operate by some pretty rigid ex-ante rules about how they can inspect people. On the other hand, Israel invests a lot of money on their airline screening. So this is a famous story that Katherine knows about. I once came into Ben Gurion Airport for a conference, and I had a Libyan multiple entry visa in my passport because I had been talking to Colonel Gaddafi. And so I go up to the security guy, hand him my passport. He says, 'Ah, so you've been to Libya. What were you doing there?' I said, 'Well, I was talking to Colonel Gaddafi.' And he looks at my passport. He looks at me. He says, 'Okay, you can come in.' So you would never in a gazillion years allow a high school graduate TSA employee to make that kind of a decision. But Israel, you know, invests a lot of money in having screeners who have enough background and contextual knowledge to make that decision. So in any event, and this also explains, you know, why you think about something like the Federal Reserve. The Federal Reserve has unbelievable discretionary power. During the financial crisis, they'd say, 'Okay, we're going to bail out this bank and this bank, but not this bank,' involving tens of billions of dollars, you know, in those kinds of decisions. And how can you do that? How can you do that safely? Well, basically the entire staff of the Fed are PhD economists, and they have a lot of background, a lot of training. They're not high school graduates, right? And so this was the basic theory that the appropriate degree of bureaucratic autonomy should be determined by the underlying degree of capacity, meaning basically education, skills, and then there's an ethical element to this because I think a professional education, among other things, provides you with a set of norms where you're not just serving yourself, you're also trying to serve a, you know, a larger social purpose, and that is a framework for understanding how much discretion you should provide. Okay, this is a little bit of a diversion, but it still explains some of the stuff that I wrote.
When I arrived at Stanford in 2010, we were, I think, kind of at the tail end of this period that had seen the rise of new institutional economics, right? This was Doug North and all of his followers. And so, you know, there's a whole series of writings where economists had not paid attention to institutions prior to this. Seems to me kind of unbelievable that they didn't, you know, that politics didn't enter into their basic growth models like Harrod-Domar and so forth. But, you know, North really changed that, and so they began to write about the importance of institutions, political institutions. So you've got, you know, Acemoglu and Robinson, and, you know, a lot of people, Barry Weingast here at Stanford, were writing in this tradition. So I thought this was great. You know, they've discovered institutions, politics is important. Things like rule of law are important, but it also kind of drove me crazy because all of these economists kind of were trying to reinvent the wheel. So they come up with all these neologisms like inclusive institutions, open access orders, that actually were used to describe things that political scientists actually had a vocabulary to describe for many, many years. And in particular, they never really, I think, specified what kinds of institutions were important. So if you go back to Doug North, it's all about property rights and contract enforcement. If you get to Acemoglu and Robinson, it's all, you know, it's kind of a mixture of property rights, contract enforcement, but also democracy, you know, rule of law, all these other good things mixed in. And I think that what that literature missed was that the actual institution that is really important if you want to develop economically is a modern state, a state that has high capacity, that has kind of meritocratic recruitment and so forth. And, you know, my evidence for that is they just don't have a way of explaining, you know, Acemoglu and Robinson can't explain China. They can't explain why China became such a rich country over this long period of time, lacking any kind of democratic accountability, lacking anything remotely like a Western rule of law. But they did have a very competent modern state, and, you know, that's what allowed them to develop as quickly as they did. All right, that's kind of a... Okay, so now we're going to move to contemporary America. This question of delegation is absolutely at the core of our current controversies between red and blue, between Trumpists and Democrats, you know, right now. So delegation has been a really sore point among American conservatives, particularly since the New Deal, since the 1930s. So all these complaints about the administrative state really date back to Roosevelt's expansion of the American administrative state in the 1930s, with all these new agencies and ministries and cabinet departments and so forth. And the line that you hear from conservatives, you know, this came out of Elon Musk's mouth, you know, in the last few weeks, is that the bureaucrats are unelected. They are out of control. They're making all sorts of decisions that escape the control of democratically elected political officials. Have you heard that before? Right. It's a pretty common thing for conservatives to say. And furthermore, bureaucrats have these left-leaning biases that are out of touch with what the American people want. I mean, the last time I checked, most policemen and firemen were not left-leaning, but, you know, there is something to that. There are, you know, you go to the EPA or the Education Department, that's probably true, but I don't think it's a universal characteristic. But the general idea was that the democratically elected legitimate leaders in the United States could not control the bureaucracy and that we were being governed by a bureaucratic tyranny. And, you know, there are various constitutional glosses put on this. There was supposedly a non-delegation doctrine that somehow conservative jurists found in the Constitution. This is nonsense. You know, the first Congress of the United States delegated to Alexander Hamilton the responsibility, you know, his Treasury Department for cleaning up the Revolutionary War debt. Right? So this is something that's existed in American constitutional practice right since the beginning of the republic. Okay? But that's been the argument. Whoops, sorry. So there's an empirical question. To what extent are bureaucrats actually out of control? I don't think that anybody has tried to study it systematically, empirically, and it's hard to do because, you know, there's such a diversity of agencies and so many decisions made over time. But you can certainly pick out individual cases where bureaucrats have exceeded their authority. So, you know, there's a decision that was actually just revisited by this Supreme Court, Sackett versus EPA, in which the EPA said that they have authority over American waterways. And they basically told a property owner in a landlocked part of Oklahoma that they couldn't build on their property because a bunch of seabirds might be able to use that property to land on their way to actual water. And they said, 'Okay, that means that this is actually subject to the authority of our control over the waterways of the United States, even though there wasn't any water on this property, right?' And Sackett said, 'What the hell?' you know.
And they actually won that case. Loper Bright — this was the one that overturned Chevron deference. That was a case where, I forget whether it was the EPA, some other agency had declared that fishing boats had to carry certain equipment in order to comply with some environmental rule. It was an unfunded mandate. They took it to the court and the court said, 'No, you know, you're exceeding your authority by doing this.' Title IX. So, all of you who have taken a sexual harassment, sexual assault course — we used to do it every three years, now we do it every year. This is a result of an expansion of Title IX authority that was done by the Office of Civil Rights in the Education Department. In my humble opinion, it should have been put through an APA process, an Administrative Procedure Act with notice and comment, but you know, they got around that and so they forced, you know, compliance with something that I think was actually not really in the authority of the Education Department to do. So there are definitely cases where you have bureaucracies that are making decisions and making policy basically without going through the proper channels.
So, Kate Bers and I — so Kate was a postdoc here at CDDRL about 10 years ago. She's now teaching at Davidson. But we wrote an article — Margaret's not here, huh? So Margaret allowed us to write an article on bureaucratic autonomy for the Annual Review of Political Science. And this was really a fascinating exercise for both Kate and me because we're both political scientists and we really had no familiarity with administrative law whatsoever. And so we both had to do kind of a crash course on American administrative law. And what we argued in the article was it is not true that bureaucrats are out of control. And in fact there are a lot of control mechanisms that exist by which the political layer controls the bureaucratic layer. Right? So you've got ex ante rules. For example, the Federal Acquisition Regulations are ex ante rules for how you procure things. If you are a federal agency, this rule book extends to several thousand pages of very detailed instructions. This is the reason why it costs so much for the federal government to buy a hammer, you know, or an airplane or a, you know, office computer, or whatnot, right? So, there are a lot of ex ante rules in personnel, in procurement, in lots of other areas of government. You have ex post accountability. So, you probably noticed that Congress is constantly calling agency heads and bureaucrats to the carpet. You know, 'How did you manage to excuse that bank? You know, didn't you notice criminal activity going on?' So there's ex post accountability. Then the two important ones are appointment power and removal power. So obviously the real way that you control a bureaucracy is by who you get to run it.
And here there's actually a constitutional issue. So the Constitution gives the president the very clear ability to appoint senior officials with the advice and consent of the Senate. Right? It does not say that much about inferior officers, whether Congress has the right to regulate the appointment of inferior officers in an American bureaucracy. And then removal power, you know, this is the ability to fire people. Like this is completely at the center of the controversy with DOGE right now because Congress actually has passed many, many rules about removals. So there are all sorts of civil service protections that begin with the Pendleton Act in 1883 that control hiring, but it also makes certain positions subject to what they call for-cause removal. So the head of the BLS, the head of the IRS, many agencies cannot be fired by the president without for-cause justification, meaning you basically have had to assault a kid or something, you know, before you can be removed. You have to give, you know, 60 days' notice and you have to explain why you're removing these people. So, the Trump administration and Elon Musk have basically fired all of these people. Inspectors general, you know, there's several hundred positions throughout the federal government that have for-cause removal protection. The right of Congress to set these rules was created by a Supreme Court decision in the 1930s called Humphrey's Executor. And the Trump administration came into office saying we don't believe this is constitutional. And so this is one of the big decisions that's now going to go up before the Supreme Court, I would say probably in the next couple of months, where the Supreme Court is going to revisit whether Congress actually has the right to protect certain job categories. And I, you know, my sense from the administrative lawyers I've talked to is that they'll probably overturn this and support, you know, the administration's ability to remove this particular category of officials.
There's a long history to this. So James Madison was the founding father that was the big proponent of the president's ability to remove officials. He said, you know, unlike appointment power, you know, if you want to appoint somebody and the Senate turns you down, you can always appoint somebody else, but if you can't remove an official, that really impedes your executive authority. And I think that contemporary theories of the unitary executive come out of that debate. It was called the Decision of 1789, where, you know, the founding fathers actually tried to hash out this issue, and the law on this is still a little bit unclear. But removal power for senior officials — absolutely nobody questions the ability of a president to do that, and it's a very important way that elected political officials have of disciplining the bureaucracy. And then finally, ad hoc interventions, emergency powers. You know, you can override rules in certain cases. The California state constitution gives the governor the ability to declare a state emergency. So Jerry Brown actually did this in 2015. He said that the drought in California is giving him — he said that that's created an emergency situation where he could suspend CEQA. This, you know, this law — you've probably heard me complain about this in the past, right? So there are things of that sort, right? So that's where we kind of are in American law, where you've got an administration that has basically come into office saying we believe in the unitary executive, we think that the president should be able to fire everybody and anybody that that person wants. And that's really what's before the courts right now. But the point of this slide is to indicate that this argument that the political layer has no ability to control the bureaucratic layer is just nonsense. There are all these mechanisms. And I think the real problem is that Congress doesn't exercise its constitutional authority to, you know, to control the bureaucracy. That's the fundamental problem. All right.
So, so what's the real problem? So I started this working group a little over a year ago. We originally called it Protect and Reform Our Public Service. This was before the election. So, we figured we would protect the public service if the Republicans won and we would try to reform the public service if the Democrats won. We actually didn't — we knew that there was Schedule F and all these, you know, the Project 2025 attempts to attack the bureaucracy. We didn't quite grasp how pervasive and powerful that would be. But we are still at work and we are now actually in the reform wing of what we wanted to do because, you know, essentially we can't protect the bureaucracy. You got to be a lawyer and you got to be litigating and that's not our function. But we do think there's an opportunity because I think that the DOGE attack on the federal bureaucracy is going to run its course. They're going to do a tremendous amount of damage. They're going to fire a lot of people that shouldn't be fired. I think Americans are going to realize that actually the government isn't their enemy. And when they can't call their Social Security office or they can't get their veterans' benefit, they're going to say, 'Oh no, well maybe the government actually does have a purpose after all.' Right? And so I think that we are hopefully going to move to that kind of point. And actually the destruction that's been done up till now may in a way clear the decks for a real reform of the public service, and I think that we need to do that. And that requires properly understanding what is the problem with the federal bureaucracy today.
I agree with Donald Trump and Elon Musk that there is a deep problem in our federal bureaucracy. I don't dispute that it's not a well-tuned, efficient machine, right? But I think that they are basically 180 degrees wrong with what the basic problem is. Conservatives say the problem is it's out of control, that they're making all these decisions without any kind of democratic oversight. I think the actual problem with the bureaucracy is the opposite. It is over-constrained. So there are way too many rules that every single federal agency has to comply with, and it creates a culture within the bureaucracy where rule compliance becomes the way you get promoted. It's your major incentive to move forward in your career, actually producing results in the real world for your constituents — that's not what you're either punished or rewarded for. You are punished or rewarded for your ability to comply with rules. And that is really what drives people crazy about bureaucrats, is that, you know, you're at the DMV. It's 4:59. You finally get to the front of the line. And the bureaucrat says, 'Oh, sorry. It's 5:00. Time to go home.' You say, 'Yeah, but I'm the only person in line. You know, can't you just get me through?' 'Nope. Sorry. Rule says we close at 5:00. That's it.' Right? That's what drives people crazy about bureaucracies. That is not the bureaucrat's fault. That is the fault of the people that put these extensive mandates on the bureaucracy.
So Dan Ho at SLS has this project to use big data to try to actually catalog these rules that affect federal agencies. And he was telling me that they're trying to count the number of congressionally mandated annual reports that federal agencies had to submit to Congress. And he said that they can't actually come up with a number, but it's over 10,000. Right? He said that there's a report on the dollar coin that the Treasury Department is required to file every single year. And so if you want to ask what are bureaucrats — what makes them so inefficient? Well, one of the reasons is there is some bureaucrat in the Treasury Department that every year has to draft a report to Congress on what's happened to the dollar coin in the last year, right? 10,000 of these things. So, what do you expect bureaucrats to be doing, you know? Plus which there are a lot of informal rules that they think they have to follow. So this brings us to what I think a real reform of the bureaucracy would have to look like. They need more authority and not less. Kind of the opposite of this conservative argument. They need to be given more discretionary power to fulfill the legitimate mandates that Congress places on them and to use their own judgment as to how to do this best. They need to apply, you know, what's called agile principles so that they are more flexible, able to get past stupid rules and, you know, modify things on the fly.
This is something — so say a couple of MIP students in here — so I've been trying to beat this into the heads of all of our students at MIP: you cannot separate policymaking from implementation. That is a big problem in the American bureaucracy. The prestige lies in policymaking. 'I'm the one that came up with SNAP benefits that's going to help poor people across the country. But we're going to hand the implementation of this off to a contractor.' And that is a really big mistake because you can't determine what a good policy is if you don't actually get into the weeds of implementation. And in general, we got too damn many contractors in the US government. You know, we began outsourcing big time in the 1980s, you know, when Reagan was president, because they figured the private sector can do things more efficiently. Well, that's not so true. I mean, private contractors have two principles. One is the people they serve and the other principle is their shareholders. And those sometimes come into conflict. And we got to bring more capacity back into the US government. I would just recommend the best single book I know of that talks about the deep problems in the American bureaucracy was written by Jen Pahlka, who founded Code for America. She's the founder of the US Digital Service. If you read that book, you'll see exactly what the problem of the American bureaucracy is because it's just over-constrained and it needs to be liberated. All right, that's it.
