Donald Trump has taken extraordinary actions to redirect or ignore congressional appropriations, from dismantling foreign aid to making the Education Department a ghost town. But how unique are Trump’s efforts to avoid spending when he does not favor it? Kevin Angell compares what Congress appropriated to what agencies actually spent over decades, finding that presidents have long moved spending toward their preferences. Even after impoundment controls, presidents found ways of not spending money they never requested from Congress. Trump is more blatant and could be more extreme, but administrations have long used their tools to obtain some power of the purse from Congress.
Guest: Kevin Angell, University of Chicago
Study: Who Has the Power of the Purse?
We welcome your thoughts on this episode and the podcast as a whole. Please send feedback or suggestions to scienceofpolitics@niskanencenter.org
Transcript
Matt Grossmann: Who now directs spending? Congress or the President? This week on The Science of Politics. For the Niskanen Center, I’m Matt Grossmann. Donald Trump has taken extraordinary actions to redirect or ignore congressional appropriations, from dismantling foreign aid to making the education department a ghost town, to offering employee buyouts across government. But how unique are Trump’s efforts to redirect spending toward his priorities and avoid spending when he does not favor it?
This week, I talked to Kevin Angell of the University of Chicago about his new paper Who Has the Power of the Purse? He compares what Congress appropriated to what agencies actually spend, finding that presidents have long moved spending toward their preferences. Even after impoundment controls, presidents have found ways of not spending money they never requested from Congress. Trump is more blatant and could be more extreme, but administrations have long used their tools to obtain some power of the purse from Congress. I think you’ll enjoy our conversation. So tell us about the findings and takeaways from your new paper Who Has the Power of the Purse?
Kevin Angell: The main finding from my new paper is that presidents are able to influence government spending in a rather substantial way after they have agreed to a budget with Congress. So after they’ve done their negotiations with Congress, after Congress has passed the budget and the president signed it, presidents are still able to shift government spending so that it more closely reflects their priorities than is reflected in the original budget.
Matt Grossmann: And give us a sense of the size of these effects and how much they matter in terms of the actual policy outputs of government or potential outcomes from policy differences.
Kevin Angell: We see that the power is really concentrated in presidents trying to spend less than what Congress has passed and that these effects are pretty substantial. So for every about a dollar change in the President’s budget, we see about 40 cents less in spending that’s occurring in that year by that government subunit, which you can think of as a subset of an agency.
Matt Grossmann: Just to clarify, the president proposed $4 of spending, Congress proposed $5 of spending, and the president is able to move the final spending about 40% less than the difference between those two, that dollar, is that right?
Kevin Angell: Correct. We can think about this as, suppose we have two budgets, both of which Congress has passed $5 of spending. In one case, the president had wanted to spend $4 and the other case the President had wanted to spend say, $3. We’d expect that the president and his administration is going to spend about 40 cents less in the case where they asked for $3 than when they had asked for $4.
Matt Grossmann: Great. Give us some, I guess, more direct examples here. Where are these agencies differences between the president and Congress most common? Are there quintessential cases that come to mind where we can say there was a fight between Congress and the president, Congress won, but in the end, the president was able to get more by a couple of years later?
Kevin Angell: I think that one of the quintessential cases of this is actually slightly before the time period that I study, and we actually have really good evidence from our qualitative history about this happening. And that’s with the Truman administration. In aftermath of World War II, Truman and the House and Senate were in a dispute over how big the Air Force should be. The House, in particular, wanted a substantially larger Air Force than President Truman did.
And the way that Truman and the Congress resolved this is that Truman basically goes to the Senate, which shared his priority to have a smaller Air Force, and said, “Look, let’s just agree to the House number where we’ll accept the extra money that they’re going to appropriate for us, and then I’ll have a handshake deal with you, the Senate, that I’m just not going to spend it.” So what happened is that Congress passed the higher budget, President Truman signs the budget and then immediately places all of the extra funds into what he called reserves so that he didn’t ultimately have to spend it on the Air Force.
Matt Grossmann: And what about the contemporary mechanisms for this? Presidents maybe. Maybe we’re back to being that abrupt, but for a long while, I imagine that this was a little more hidden, or at least in the weeds in terms of hiring or contracting or materials cost. So what do we know about the actual mechanisms for this and how much is from the agency versus directed from the Office of Management and Budget of the White House?
Kevin Angell: First, what I’ll say is I’m going to bracket the last, call it, six years or so. This paper’s data ends in 2018 so I can’t speak to the current moment about exactly how it’s happening. And in many ways, I think as we’ll talk about later, things have definitely shifted in terms of the opaqueness and blatantness that this is occurring. But since the Impoundment Control Act in the 1970s, things have really shifted, so they are more opaque. Presidents haven’t been able to clearly and publicly give these direct orders to not spend funds. And that’s kind of been considered, based off of my interviews and my data analysis, to be the red line where presidents can come up with all sorts of other ways to justify not spending funds, whether that be program timing, whether it be changing paperwork, but they couldn’t say, “I’m doing this to not spend this money,” even if that might be the ultimate end result of that action.
So until the present moment, you’re right, presidents haven’t been able to say the goal is to impound funds. They look for other reasons to justify those changes. In terms of how this is happening, this relates to some work that I’m currently doing trying to disentangle whether this is changing in awarding the contracts or changing in the number of personnel that are employed. Those are very disparate data sets, and I don’t have final results to share there. But based off of some interviews with practitioners, this seems to be a product of, one, there’s a lot that has to do with administrative burden and administering these programs where presidents and their agencies can strategically change how much burden there is. But then also in terms of whether this is coming from the president or not, I’ve chatted with some folks, including some folks at independent agencies that say ultimately while they’re in control of it, they might and do definitely take into account the preferences of those who may have appointed them or otherwise influencing the budget when they’re making decisions about how much to spend their budget.
Matt Grossmann: So you can, of course, imagine innocent explanations for this underspending. “We were trying to hire someone, they didn’t accept our job, we contracted, but the contractor didn’t finish the work within the year,” those kinds of things. How much is there assigned that these kinds of explanations are partially responsible versus their excuses?
Kevin Angell: You’re totally correct that they’re all, like you said, innocent explanations for this. However, in my approach, what I’m doing is I’m trying to take account for those and look at changes that are clearly correlated with changes in how much presidents propose. Yes, there might be cases where the bridge goes over budget or maybe you get a great deal on your bridge contract and you come in under budget, but we shouldn’t see those only happening when presidents want them to happen.
If presidents don’t want to spend funds in a particular area, say public assistance, and what do you know that year there aren’t mysteriously the number of people who apply for public assistance drastically falls, that starts to look a lot less like random chance and more so like active implementation changes. I’m fairly confident, especially given the size of these effects, that this isn’t just due to one-off innocent impacts.
Matt Grossmann: But we might imagine that the president’s going to appoint people to the agencies where he wants to see cuts that also don’t think the money needs to be spent or that the budget could signal the kinds of priorities of the people in these agencies. I don’t know, how much evidence do we have that it really requires intervention rather than anticipation or the personnel at those agencies?
Kevin Angell: I think this is a really good point, Matt, and this is something that my paper doesn’t disentangle between, whether this is anticipation of a presidential directive or whether this is because of behind-the-scenes directed from the president. If anything, I perhaps myself would lean more towards this is due to anticipation or because presidents have appointed those who share their ideology. But if you think about the classic principal agent framework in which presidents are trying to exercise influence in an agency, whether or not the president is directly telling the head of an agency, “Don’t spend this money,” or is going and finding someone who is just going to not spend that money on their own, that might shape our understanding of exactly what we should expect and where we should look as we’re trying to investigate this, but it doesn’t change the overall effect that this money isn’t getting spent when Congress has instructed that it be spent.
Matt Grossmann: Regardless of the mechanisms or even the president as the cause here, it seems like this should really change our theories of how the president and Congress come to consensus on appropriations about bargaining, about the role of presidential power. How much should this change our theories of how the president and Congress interact, and does it explain any anomalies in budget deals that we’ve seen, maybe similar to that example that you mentioned about the Air Force?
Kevin Angell: Sure. I think one of the big ways that this should shape our understanding of congressional bargaining and presidential power is that the bargaining process doesn’t simply end with the adoption of legislation. And this is something we’ve seen pieces of in previous research about how policy decisions, whether that be where grants are made or who gets grants, that after the legislation has passed in Congress, all of a sudden the lobbying all shifts to the executive branch. So it goes from the halls of Congress to maybe the halls of the EEOB, the Eisenhower Executive Building, where a lot of the OMB staff sit. So I see this paper as really reinforcing the fact that as we’re considering who wins in the legislative bargain, we also need to take account who is able to win in executive implementation because the person that might look like a winner upfront might end up losing when it comes to the White House and the person who lost in Congress may be able to get their second chance by lobbying the president and the staff.
In terms of how this shapes our other understandings, one of the things that I really want to focus on is why this is occurring. You can think of three main reasons why Congress would allow this to happen. The first is that Congress just doesn’t have the information it needs at the time that it’s making its decisions to know about this. So this could be Congress doesn’t really know that this is happening, this is happening behind the scenes, and that’s why it’s continuing. The second idea might be that Congress does know about it and that they’re fine with it. You might imagine that Congress knows that they’re going to have to pass some things that they don’t particularly like, maybe some pork that they think is wasteful or whatnot, but they’re relying on the president to really exercise his management authority and cut those out so congressmen can go home to their congressional district and say, “Look, I got this pork for you. Go blame the President, he’s the one who cut it.”
And then the third one would be that Congress does know about it, but they’re unable to solve the collective action challenges that are necessary to enforcing their own budget legislation. This is the area that I think is probably perhaps most likely is that Congress faces this big issue, it’s got its committee system, it’s got all sorts of other areas where Congress is splitting its oversight. The people who might be most interested in the funds aren’t the ones with the best access to information about whether it’s spent or maybe they are, but they have different preferences in the rest of the chamber. So I really think that all those accounts can be really informed by this. And this opens up a lot of new lines for inquiry.
Matt Grossmann: It seems to explain some regularities that maybe seem odd without knowing yet. One is that the president submits a budget and Congress completely ignores it. We also have this idea that the appropriations committees are extremely powerful and usually get what they want. And yet we have the anomaly that people in the administration certainly spend a lot of time preparing the budgets for the president’s budget proposal. So maybe this does connect those together to say everybody is sort of acting in line with the findings. The presidential budget process is maybe not as much for winning in Congress, but for preparing afterwards. What do you think of that story?
Kevin Angell: When we talk about the government budget, I do want to be a little bit precise here. The president submits, at the start of every year, a budget that’s in compliance with the 1921 General Accounting Act, what we usually call the Budget and Accounting Act of 1921. And then after they deliver that to Congress, Congress takes it and has a whole bunch of different appropriation bills and the usual process. Sometimes you get a nominal bust like we’ve had in recent years.
But what I think is important here is that, yes, in theory, Congress takes the president’s budget and throws it away. It’s actually not what we necessarily see in the data. There’s lots of cases where the number that comes out of Congress is the number that came in the president’s budget. So now whether that’s on important issues or not, I’ll leave individual readers to think, everyone has something they think is important. But empirically, we don’t see the budget being in Congress starting from scratch from the budget that the president submitted. That’s one of the reasons why the president is required to submit the budget.
… president submitted. That’s one of the reasons why the president is required to submit the budget, because this is an enormous task that Congress at the time didn’t have the resources and now probably doesn’t want to use its resources on producing that document from scratch.
In terms of the irregularities, I think what this really can highlight is the fact that the people who have the expert knowledge in this are going to be in the classical setting your committees and your committee members, and so those are going to be the people that are best empowered to cut those side deals with Congress because they’re the ones that are going to be exercising oversight of this moving forward. And so one area that I’m working on right now is looking at, are the people who have better access to resources in Congress, namely that committees, better able to see their priorities continue to be reflected in presidential implementation of the budget?
Matt Grossmann: So I know we don’t want to get too far into the weeds of your adventure collecting all of this data, but it sort of helps to explain why journalists and members of Congress maybe haven’t been as on this as you might expect. So how did you go about doing this? Why does it take two years to wait to get a final number? And how many people do you think have seen these discrepancies before and know about this process, at least within their agency or domain?
Kevin Angell: Yeah, so in terms of the process, what I did is I went back to those budget documents that presidents propose every year. We have them going back for decades. So I went and I found those scans, and for every subunit, so these are units inside of agencies, I went and I found how much the president asked for in that year’s document. I did that for all the agencies, and then I would go two years later when the president is required to submit, how much the budget actually ended up being and then how much the budget was actually spent. For each of those agencies, I gathered that itself and then I linked them up. And so all this data is coming from the budget. So in theory, anyone could have gone out and gathered this and could still do it, but especially when you’re trying to do it over multiple years, it just takes a long time. Like you said, presidents only submit the actuals about a year and a half later, so for the, say, 2022 budget, we’ll get the actuals in the 2024 budget.
And so at that point, may have had an election cycle, the people who were in Congress, when the budget was passed may be different, they might have different priorities. And so that lag I think is an important factor in explaining why we don’t have as much policing on this because Congress doesn’t have its own ability to do all its own accounting. It’s relying on, in many ways, OMB and other agencies inside the executive branch to do that accounting.
Matt Grossmann: And what about the second part? Is this kind of a known thing within these agencies when they’ve underspent and they know it before those final numbers are submitted, or is this kind of opaque to everyone until later?
Kevin Angell: I think the agencies would have the ability to know it themselves. Whether they know the exact number is going to depend on some of the timing of contracts at the end of the year. But I think it’s pretty fair to say that the agencies will have an understanding that they are under budget, especially in cases when… I will say agencies are much more worried about going over budget than they are under budget. Going over budget has much more severe penalties laid out in statute than underspending. And so my understanding from interviewing agency staff is that going over is their focus. They don’t mind going under, especially if they’re not going to face congressional repercussions, but that essentially do or die on making sure not to go over.
Matt Grossmann: So we know that the basic pattern is that these cuts happen most when presidents proposed underspending and that underspending is way more common than overspending. What are the other big variations? Were there some presidents that moved this a lot more than others? Are there certain kinds of budget accounts or agencies where this is easy?
Kevin Angell: The primary difference that I really see is between discretionary and mandatory spending, and I think this is something that isn’t necessarily obvious that it should fall in one or the other. So on one hand, discretionary spending, with its usually annual appropriations process, might have greater congressional oversight. It’s the Appropriations committee fighting for what the budget is each year, and they know what was spent last year; they’re reviewing the receipts as they’re moving forward. And so we might think that discretionary spending, this would happen less. On the other hand, we could make the same case for mandatory spending where Congress sets the eligibility formula for a particular program, and then the president goes off and implements it and is able to make all sorts of adjustments on exactly how to implement that without the same sort of annual oversight process. So conceptually, it’s not quite clear which one we should expect the most in.
What I see from the data is that very strong, I’d say… We see strong evidence that it’s predominantly focused on discretionary spending, so it’s inside of discretionary spending that presidents are able to exert this influence, whereas we don’t see that in our mandatory spending programs.
Matt Grossmann: You explained a bit that it is much more common to see underspending than overspending. Give us a little bit of the reasons for that. Are these political constraints? Are they just kind of administrative patterns? And why doesn’t this give an inherent advantage to conservative presidents? Why don’t we see the big changes being Republican presidents facing Democratic congresses? Is it because they don’t really have the differences in priorities that we expect, or is there some other explanation?
Kevin Angell: Yeah. So in terms of why we see differences in overspending and underspending, the first thing we want to highlight is that the two operate under very different legal regimes. Overspending is governed by what’s known as the Anti-Deficiency Act, which actually carries both criminal and administrative penalties for violating it. So if you’re involved in spending money without congressional authorization, spending money that shouldn’t be spent, you can actually be prosecuted for that. Whereas if you don’t spend money that has been authorized, you might get an Impoundment Control Act investigation initiated against you by GAO, but there’s no clear penalties for that.
And so one thing that I can highlight is at the heart of the first Trump impeachment, which we cannot classically hear about it being the Ukraine scandal, at the heart of that was an Impoundment Control Act violation. Money was supposed to be spent that wasn’t. Ultimately President Trump was impeached for it, though he wasn’t convicted. So if you’re a civil servant, that might be a much more remote penalty to you than if you authorize an overspending and you potentially get criminally charged for that. So that’s kind of one reason why I think that this predominates and can be reflected of this.
In my conversations with staff, OMB systems, agency systems are all set up to be very, very strict about overspending, so if you try to go in and spend money that your agency doesn’t have, in some cases it’s quite literally hard-coded you can’t do that. In other cases, it sends up all sorts of alarm bells. Whereas there’s not the equivalent on the underspending side. It’s not like there’s an alarm bell that goes off if you spend 95%, not 100% of your budget. And if there are controls, it’s very much weaker than preventing you from putting in that payment that is going over your budget.
In terms of how this plays out partisan-wise, I do want to make a distinction between the ability to influence and how much the presidents actually use that influence. So seeing the data is that when presidents tried to underspend, presidents are equally successful regardless of their party. So Democrats have the same ability to underspend as Republicans do, but sure, we see Republicans do this more frequently. They want to spend less money in general, at least that’s what classical understanding says, and so they in fact do propose more spending cuts and implement more spending cuts.
But I think the one reason why we don’t see this massive focus on it’s only Republicans is that everyone can find some government spending that they don’t like. It might be in different areas. Using the contemporary situation, perhaps Democrats want to spend more on public assistance, but they want to spend a whole lot less on, say, ICE. Whereas Republicans might be the reverse. So everyone likes their own pot of government spending, but that means they equally don’t like other areas. You can see this in the Department of Agriculture, this huge department, and I look at some pieces of it, but it encompasses everything from food assistance for needy Americans to crop subsidies to farmers. And so you might think that even inside what we look as, oh, agriculture spending, you still have parts of it that you like and parts of it that you don’t.
Matt Grossmann: So just to clarify, Republican presidents overall do request bigger cuts compared to congressional appropriations; they just move it about as much in the same direction as Democrats. Is that right?
Kevin Angell: That’s correct. And so one other thing to add here is that, why don’t we see Republican presidents maybe taking this to the limit? The classic understanding is that not giving people funds is likely going to perhaps be more painful than if everyone gets extra. If you think about your administering a grant program and you spend over, an extra organization gets a grant, you’re not necessarily going to have as many people calling and yelling at you than if you were to tell a bunch of organizations, “Actually, this year, you don’t get your funding.” And so there are these natural factors that influence underspending and overspending. Underspending might get more angry phone calls, overspending, harsher penalties, and at some point there’s just not money left in the bank.
Matt Grossmann: So the folk history of this is that it was kind of going on a little bit for a while, but that Nixon really took it to an extreme, and then Congress responded to that with a policy change that had more teeth. How much does that really show up in your data? And if this wasn’t a solution, why wasn’t it? And is this just a story of the continuity is there and just the legal justifications and routes change, or are there real breaks when we notice?
Kevin Angell: So the first piece of that that I’ll tackle is, what is different about Nixon? So if we look at the different breakdowns by president, Nixon doesn’t really stand out as particularly an outlier in how much funding he’s influenced. But what you do see if you read some of the more qualitative accounts is that, one, that the type of program that he goes after is different than what happened in the immediate past. So prior to that, a lot more had been in your defense or national security areas, whereas Nixon really drives it home with domestic assistance programs, most notably, say, urban renewal or urban development. So you’re right that Congress does come out and they pass the Impoundment Control Act, but I really want to highlight, it’s actually a cluster of three reforms and cases that really could be doing the work. And so in my paper, I don’t highlight on any specific one. I just do, “Look, we have these three reforms that happen over these two years, and what we can see is before and after that.”
So we have the Impoundment Control Act, which tries to create this categories for how presidents can propose deferrals or rescissions. We have a case called Train v. New York, which is a litigation about what existing statutes allowed the president to do. And Train v. New York, it said that presidents were exceeding their authority under existing statutes to impound funds. And then the third one is there’s a revision to the Anti-Deficiency Act, which Congress intended to pass to prevent presidents from overspending, while also allowing them to set aside funds to account for things like inflation or changing circumstances. But that gets really tightened in the 1970s because previously presidents had said, “Look, we know you passed this law to keep people from spending too much, but one way we can keep from spending too much is we can decide that we don’t want to spend any of this,” and Congress was reacting to that misuse in their view of the Anti-Deficiency Act.
In terms of how this changes before and after, you’re right, it introduces a whole new regime where now presidents are required to follow these set procedures when they want to not spend funds, but I want to highlight, the Impoundment Control Act isn’t that clear of a law, and there’s not clear enforcement mechanisms. Like I mentioned earlier in the interview, there’s no criminal penalties for violating it. It’s also not clear what in fact an ICA violation is. I’ll defer to the lawyers on this; there’s a whole history there. But some basic facts that we think we know. Does the Comptroller General have the ability to sue in federal court to enforce spending? The statute gives them that power, but that ability and the constitutionality of that has never been tested. The Comptroller General has only ever filed suit once and dropped that suit because the administration released the funds before it carried forward.
And so there’s a lot of uncertainty with how this is going to come out if this is litigated. There’s categories like programmatic delays or deferrals that kind of are created by GAO to classify things that are these innocuous reasons why money isn’t spent, but those don’t have a clear definition in statute; it’s not in the Impoundment Control Act. And so what you’re seeing now and what you see after the ICA is passed is that presidents are looking and poking around and trying to find the holes, and I would say my paper is-
… holes and I would say my paper is evidence that they have been successful in finding holes.
Matt Grossmann: And just to clarify, did it take them a while or do we not really see any difference in terms of Ford and Carter and Reagan? They were all able to kind of influence appropriations right away?
Kevin Angell: Yeah. So if I make a little bit more assumptions in the data analysis process, I am able to plot a yearly estimate of how much influence presidents have. And so what you do see there is it looks like there’s a pretty sharp drop right around the Impoundment Control Act, the ADA revision and Train vs. New York, although it rebounds afterwards. So I want to be cautious about not over interpreting that, but I think it’s suggestive evidence that it at least did take the presidents a little bit of time to figure out how to operate under the new regime.
I will say one of the surprising things and that I’m currently investigating right now, is why does Nixon sign the Impoundment Control Act? There’s actually a famous political cartoon where Nixon has three henchmen that he’s using to control Congress. One is the veto power, one is executive privilege and one is impoundment, yet Nixon rolls over quickly and signs the ICA. So it makes you wonder what was going through Nixon’s head about why he allowed this. Did he just think, well, I can survive the formal restriction by coming up with all sorts of new means or what was happening?
Matt Grossmann: A lot of people think we’re at another big break in our time series with the second Trump administration. I know that you have no data from the new administration and you have only a couple of years of data from the first Trump administration. But I want to talk a little bit about it, at least let’s just start, theoretically.
One story is that this is just an administration with very different preferences. So these budgets, and particularly Russ Vought’s vision of government is just very different. So even if Trump was only able to influence spending as much as previous presidents, he just starts with a very different preference. So 30% or 40% of his moves would just be huge moves. How should we think about that, sort of the breadth of the vision here versus the abilities and tools?
Kevin Angell: I think that there’s kind of two areas to focus on here. The first one is the openness with which it’s being done. And the second one, and I’ll bracket that for now, is the scale at which it’s being done on. The first thing with openness, the previous rule of thumb that had existed prior to the first Trump administration that actually some things happened related to it in the Biden administration, was that if you were trying to not spend funds, you didn’t come out and say it, because if you came out and say it, that would be a clear trigger for a GAO investigation. Now all of a sudden you have the president on record, or agency staff member on record saying, the reason why we’re doing this is because we don’t want to spend funds.
That makes it very hard to claim that actually no, this is about some program decision that has incidental impacts on not spending funds, but is actually because of some program priority. And so what we see here is even prior to the Trump administration taking office for their second term, President Trump was pretty clear that he believed the Impoundment Control Act to be unconstitutional, that he wanted it to be challenged in court. And as we can see in Roosevelt’s confirmation hearings, he’s pretty clear that he supports the position of administration. In the administration’s opinion, the ICA is unconstitutional and they seem to be eager for a challenge to it. And so I think part of this is trying to push the limit because they want that lawsuit. They want that Supreme Court decision that strikes down the Impoundment Control Act, and this is their way of triggering that.
In terms of the scale, I think one of the pieces that really you see openly here that you didn’t see before is in personnel, which opens a whole nother can of worms because personnel politics are very complicated, you have all sorts of regulations for how and when you can fire people, under what reasons. And so that’s something that we’re seeing openly happen in the Trump administration that at least wasn’t on the surface happening prior to it, although I am doing more research to see how much was this adjustments. Although I think it’s fair to say that to the scale that we know DOGE, where the focus is on the riffing of thousands of people, I haven’t seen any instances of that in previous impoundment discussions.
Matt Grossmann: And then just did you want to address this kind of theoretical issue of how should we think about it, if it ends up being the case that Trump is no better at moving things towards his priorities, but just that they’re just so different than Congress says, that the spending reductions are going to look bigger because the proposed reductions are bigger?
Kevin Angell: Yeah, I think theoretically the question to ask here is normally, do we want this to occur and do we want presidents to be able to shift by a percentage rather than give them a subset flexibility? I think it’s conceivable that a reform might be, look, presidents have X percent on either side that they can spend as kind of discretionary funds, but that outside of that there’s stricter guardrails. And so as we’re considering what does it mean for the Trump administration to try to cut spending and should we restrict it, we have to ask ourselves normally what do we want and why? And if we want flexibility, why do we want flexibility? I don’t think in many people’s mind flexibility in itself is the goal, it’s to have some sort of better government outcome.
Matt Grossmann: So you said that there wasn’t much precedent or you couldn’t find much precedent for the DOGE or reduction in forces. I know that they also kind of canceled a bunch of contracts for future spending in DOGE. And then we have these examples like USAID where a whole agency has kind of been moved within the State Department and the Consumer Financial Protection Bureau where they tried to kind of zero out the accounts. Are there proto versions of those kinds of moves in your data or anything that we could look on as precedent for that?
Kevin Angell: Yes. I mean, there’s lots of examples of government trying to modernize and reorganization. And Congress has gone through various iterations where they’ve delegated different authority to the president to reorganize executive departments. And there’s mechanisms inside of various budget bills where you can reprogram funds that are meant for one purpose to another related purpose.
But I do think that, we haven’t seen this example where government departments have become this key focus of a DOGE-like instrument. Republicans for a long time talked about getting rid of the Department of Education, and there were lots of plans for doing so. Legislation got proposed, et cetera, but we didn’t have this DOGE equivalent that just went in and sent everyone home until the current moment. So again, normally it’s up to listeners to know whether they want presidents to have this flexibility, but it’s not clear what the limit is on that.
Matt Grossmann: So one of the reasons that they’ve been able to do more is that they’re not maybe as constrained by the courts as people expected. It takes the contractors or the people who would lose money to have standing, they’re often told to go to specific government courts to adjudicate their own contracts rather than bring a broader one. They have been winning a lot of lower level rulings but then are overruled by Supreme Court rulings. How did previous presidents think about this? Did they think there was really a court block on going too far or has this just sort of always been available and Trump is kind of the first to push the envelope?
Kevin Angell: I think that before we focus on the court enforcement of the spending, one of the things I want to highlight is one of the natural ways that Congress, or I shouldn’t say natural, one of the frequent ways that Congress has required an enforced presence to spend funds has been through a tool known as committee reports. These are essentially specific requests or directions or suggestions, depending on what language is used, made by the committee to the agency for how money should be spent. And so the classical understanding is the way that Congress would oversee if presidents didn’t spend the money in a particular area, is that if an agency wasn’t spending funds, there’d be some sort of oversight hearing or communication from the chair ranking member about, “Hey, Mr. Secretary, please, you’d better spend these funds.” And then there’d be threatened retribution against next year’s budget for that agency if those funds weren’t spent.
And so this is prior to the year even ending. You have this, and it is not really happening in the courts, it’s happening via this congressional oversight. That’s something that has changed over the years as we’ve moved away from all these separate appropriations bills towards omnibuses, exactly what committee reports are doing. What if that is adopted in, who’s conducting that oversight, has started to get a little bit murky. And so that’s the classical way this is addressed rather than through a court case. Although we’ve had our share of those, Train vs. New York, like I mentioned before, was an example of New York City not receiving the funds that it believed it was entitled to. And they did litigate that. But I think in previous years we just saw this addressed through congressional oversight rather than judicial oversight.
Matt Grossmann: And we do have a curious case here where the president, or at least DOGE wanted to tout the amount of spending that they saved, and it turned out that the numbers didn’t seem to bear that out. We also have people trying to verify the daily government spending amounts overall, and saying that not as much is changing as it might look. I know that you have to wait a year and a half to see what actually happened here, but do you think there’s a case that actually not as much is happening as people think, or do you think we are really going to see real shifts this year?
Kevin Angell: I’m not sure, Matt, and a couple of reasons why is one, we had these banner changes in how much was spent and canceling of contracts. At the same time, we did see at the end, a few different agencies rush to get money out the door towards their preferred grant recipients or their preferred programs. And so I am not quite sure how much has gone out the door, because especially if they go out the door at the end, there’s a lag for when those actually show up in the various government recording systems.
So I’ll be very interested to see, and I’ll be tracking it closely as we start to get the data, but I don’t know that we know right now at a macro level how much change we’ll see. In general, I’m a little bit skeptical that you can really turn around a government and slash spending all this quickly, especially without legislation. I mean, we did have a rescissions package that went through, but relative to the scale that the DOGE team was trying to cut, I think that rescissions package was quite small, although it was large in absolute terms. And we’ll have to see what happens as, like I said, as the data comes in.
Matt Grossmann: So yeah, we did actually pass one rescissions package and then we had one pocket rescission officially kind of move forward. And that does seem to be at least one kind of acknowledged error in the Impoundment Control Act, is that it does enable these very late requests that Congress doesn’t have time to respond to.
So let’s say that at least maybe under the next Democratic president, we do start to get some, at least behind the scenes, work on an update to the Impoundment Control Act and some effort to kind of rein in the president here. What would those kinds of tools look like or reforms look like? And does the history here suggest that this is not something that’s going to work for very long because presidents are going to find a way around it?
Kevin Angell: Yeah, I think in terms of what a reform would look like, one that I could see happening is just clearer definitions for what all the terms mean in the Impoundment Control Act. So I think there’s a broad recognition that Congress can’t say you have to spend exactly $1 billion, and if there’s one over, one under, this is some sort of major crisis. It’s hard for me to go to the grocery store and spend exactly 20 bucks, much less trying to spend $20 billion on a complicated government program. So because of that, I would expect that there’d be some sort of definition for how much leeway is okay. And then inside of that band, you’re going to have congressional oversight for, are those justifications valid? And then outside of those bands, you’re going to have much stronger judicial oversight for, okay, this isn’t just a question of does Congress buy your declaration that this is necessary, now it’s a legal consequence if it hasn’t met certain requirements.
So I would expect that to be one. I would imagine there’d be a discussion about how many votes does it take to pass a rescissions request? In some ways, when things go through budget reconciliation, the threshold is the same as it is with a rescission. It’s a bare majority, but that doesn’t reflect what’s needed in a traditional bill to overcome the filibuster. So I imagine there’ll be just some discussion over there. The reason why the ICA does privilege that bill is because there’s a recognition that Congress should vote on this before the end of the fiscal year, or at least quickly, so that the president has his instruction-
… at least quickly so that the president has his instructions on what to spend. So there could be tweaks about maybe not just they have 45 days to do it, but a vote must be held and on other forms on that front.
In terms of the effectiveness, I’m not really sure. We’d have to assess any bill that came out to really understand what effects it would have. I would say presidents are pretty protective of this. Presidents don’t want to give up this discretion and have in the past fought to avoid it.
One thing that is a possibility and that I do want to raise is that in previous instances of this, going back to as far as FDR is when I have evidence of it, when there has been concern about presidents impounding funds, presidents have been very skillful at mollifying or assuaging the concerns of the key senators to prevent them from going after discretion overall.
So there’s a famous case under FDR. FDR impounds funds for a dam during World War II. Congress starts to put the fit about this, particularly the Senator who’s concerned with this dam being built. A writer gets floated to limit on the president’s ability to impound funds generally. What do you know that the senator gets his dam and the writer disappears? So that’s a whole lot of politics there and going beyond my expertise in forecasting how the political aspect is going to turn out. But I wouldn’t be surprised if we had some sort of deal being cut to assuage concerns on a particular set of areas in exchange for there not being a wholesale reform of the ICA.
Matt Grossmann: So we had on another episode, we covered the history of war powers, and the timeline is sort of remarkably similar in that the president gains powers that were previously thought to reside in Congress. There’s attempted reforms, but they don’t really change the underlying facts on the ground. And you did title your paper about the power of the purse and kind of point to these sort of fundamental constitutional issues, so is it fair to say that this is kind of another example that the executive is kind of going to gain power over time and that kind of regardless of where the Constitution vests power, that the system has centralizing of power in the Executive Branch built in?
Kevin Angell: I think in many ways you’re right that this is reflective of a broader trend towards centralization of power in the Executive Branch. I won’t go so far as to say that’s inherent to the system that we’ve built and that in the sense that it’s unstoppable. I mean, I think one of the reasons why we’ve had a centralization of powering of the president is one, Congress has all these collective action problems that might naturally push power towards the president, but also a lot of these functions have been things that Congress has gone along with and in many cases has willingly delegated.
So take a requirement that the president be the one to do all the accounting and to present the annual budget. That goes back to 1921. That isn’t necessarily a presidential function. Congress could have established an agency that takes on that function and reports to it. We could imagine some supersized CBO that just wasn’t about legislative scoring but was actually about drafting a comprehensive budget in advance of the president doing it, or maybe the president doesn’t do it at all. And so this centralization I think is both a function of national forces, but also Congress and in many cases, the American people, or at least people point to the American people as wanting this power to shift to the president.
So everyone lines up differently. I’m a big fan of Congress when it’s writing its budget legislation to be specific what it wants, because it is constitutionally the first branch and has the ability to exercise power over the purse. But if it’s vague about what it wants or its instructions, I think we’re going to see presidents try to exploit that vagueness if Congress hasn’t been as specific as possible and doesn’t have the oversight that it needs.
Matt Grossmann: Yeah. So if you’re willing, just help us think through a little bit of this normative implications here, at least a couple of, I guess, more of the sort of straightforward arguments why this might be acceptable. First of all is that agencies just might know best, and if the alternative is we’re going to buy a bunch of computers on the last day of the fiscal year because we have to spend this amount, then maybe this is better than that. And then second, I guess, is the issue that the system doesn’t have a whole lot of built-in mechanisms for spending less. We do have out of control deficit spending relative to the amount we’re willing to tax. So is this a reasonable overall check on that?
Kevin Angell: I mean, I think one of the big concerns that I have about this process is the fact that it’s not transparent. You can go look up and see what did Congress pass for this budget and who voted against it, who voted for it, what’s the legislative history of that? But if this is happening kind of behind closed doors, it’s not subject to the FOIA process because it’s all deliberative decisions and things of that nature, or at least one could make an argument. There’s real questions about, yes, the agency might say this is best, but what went into that? Was there opportunity for comment by individuals? Is this consistent with what the legislatures intended? So I think that is a major normative concern, regardless of where you fall on whether or not we want presidential discretion.
In terms of, do agencies just know best on what to spend on? I think political science tells us that agencies likely are going to have perhaps different preferences than those who are providing their budget. You might imagine that the military is going to want to spend more money on the military and the EPA is going to want to spend more money in environmental regulation, and that might not be shared by either the president or Congress or the American public. And so yes, we can point to cases where an agency bought a bunch of very nice office chairs, I could use a new office chair myself, on the last day and say, “Okay, this is waste.” But as you try to scale up, just differentiating waste or what adequate spending is gets pretty hard.
And because let’s take an example of even something on a bridge, right? So Congress gets specific and says, “You must build this one bridge, here’s the budget to do so.” Congress is likely not legislating exact dimensions of the bridge or how good the concrete is or what color it should be painted and all these sorts of things that do have implications for how much it costs.
And so I’m in favor of Congress providing a more clear criteria to agencies for what it considers best and how it should weigh, all these different decisions. But at the end, if that decision making process is opaque, it’s hard for the American public to assess whether they think it’s done right and is a normative challenge.
Matt Grossmann: So on the other side, I want to give you a chance to address the argument that there’s a big shift in the second Trump administration. We’re descending into authoritarianism and you’re papering it over here, Kevin, by saying this is a long-term trend and that presidents have kind of always been able to do that and that there aren’t kind of real lines being crossed here. So what would you say to that?
Kevin Angell: I guess the first thing that I would say is I started this project before the second Trump administration and so this isn’t a paper that’s in reaction to it. This was long going before it. I’m going to keep studying this during this administration, during the next administration, whoever it might be. And so this is in no way, shape, or form trying to weigh in on the merits or the normative implications of this discretion.
And in my view, in order for us to make a decision about how much discretion there should occur, we need to understand how much is happening right now. And previously to, I think, the current moment in political science, a lot of political scientists just assumed that there was none or at least not ones that were big enough for us to pick up. And so I see this paper as really showing that actually no, even with the current situation, even with the Impoundment Control Act prior to the Trump administration, there were holes in it the presidents were able to exploit.
And so like I mentioned before, there are differences with the current moment or at least looks to be both in terms of size and blatantness. And we didn’t even tackle all the sorts of things about in all the DOGE rifts and the firings and the layoffs, were proper procedures being followed, et cetera. So like I said, this is in no means meant to defend any current practices. I don’t have data on them even if I wanted to, and it’s just meant to inform both the public, political scientists, and policymakers about what’s happened in the past. We can make better decisions about as we look towards reforming the process, what worked in the past. You can’t stamp something out if you don’t know if your past efforts worked.
Matt Grossmann: And is there anything we didn’t get to that you wanted to include or anything you want to tell about what is next?
Kevin Angell: Yeah. So first off, I want to say thank you to a couple of funders, the National Science Foundation, the Levin Center for Oversight Democracy, as well as the Bach Fellowship for Congressional Studies for the generous financial support. That also comes with a disclaimer that all the views that I express in this podcast are my own and don’t reflect that of any of those funders. So the information you got today is my own opinion and exactly that and doesn’t necessarily reflect the NSF, the Bach Foundation, or the Levin Center.
In terms of what’s coming next, I’m working on follow-up papers to explore a lot of the dynamics I alluded to here, whether that be if the adjustments to spending are due to personnel changes or due to contracting, how presidents are able to exercise or take advantage of this influence, whether that be through appointments or direct decisions or politicized agencies. I think there’s a lot of conflicting theories for where we should see this. On one hand, in politicized agencies, presidents might have the most ability to exercise this discretion. On the other hand, in those politicized agencies, presidents might not have any need to because if people are controlling the spending that agree with you, they can go spend it on areas that you like rather than you needing to just halt the spending altogether.
So lots in store. I’ll keep doing my interviews with practitioners and Congress and the agencies. Those got put on pause a little bit with the government shutdown, so that definitely road blocked my research for a bit, but much more of that to come.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from the Niskanen Center, and I’m your host, Matt Grossman. If you like this discussion, here are the episodes you should check out next, all linked on our website. Do presidents have the power to act alone? Will Trump have unilateral power or just pretend he does? The backstory for presidential power grabs, how the president gained war powers, and threats to democracy in the second Trump administration. Thanks to Kevin Angell for joining me. Please check out who has the power of the purse, and then listen in next time.
