Federal Money & State Influence: Navigating the Layers of Fiscal Federalism

So my goal for the next hour with you is to provide a primer on fiscal federalism. Which you can think of as the economics discipline's toolkit or apparatus for making sense of the design of federal systems. So let's dive in with some big picture facts on the US system of fiscal federalism, which we're attempting to understand.

So first is just a top line fact, there's a genuinely astonishing number of government entities in the United States, over 90,000. There's, of course, the 1 unitary federal government, 50 state governments. But also 3000 counties, 35,000 plus cities and towns, 13,000 independent school districts and 39,000 so called special districts, many of which you could think of as the providers of, for example, electrical, water and gas utilities around the country.

In addition to being numerous, these government entities are also engaged in a substantial amount of activity, so two top line facts on that. Governments adding up across federal, state and local levels, account for roughly two-fifths of economic activity within the country and employ roughly 1 in 7 workers.

One of which is yours truly as an employee of the University of California system down in San Diego. So, confronted with this kind of tremendous role of government in this relatively complex design of the system of governments, there are some questions that we should be asking. Such as how should we think about the pros and cons of the US system as it's constituted?

And what responsibilities would best be designed from the perspective of making optimal economic policy to higher versus lower levels of government?

As a next step, what I'd like to do is provide sort of a brief look at three policy spaces in which we can kind of nicely see conflicts between levels of government emerging.

Which can help us to focus our attention on the question of where we might optimally think of various functions as being assigned across levels of government. So let's start with the minimum wage. So first, four facts about minimum wage policy making in the United States. First, minimum wage laws vary dramatically across the United States.

The federal minimum wage is currently $7.25 an hour. Some state and local minimum wages are now in excess of $15 an hour. So tremendous variation in what the effective law is in various labor markets around the country. Second, congressional Democrats recently proposed a $17 an hour federal minimum wage, escalating a bit from the previous fight for 15 mantra.

Third, many cities in the state of California, also elsewhere around the West coast, also New York City, also Chicago, have their own minimum wages. And fourth, in some states, state governments have actually taken efforts to block cities from enacting minimum wages in excess of the state minimum wage.

And so you can see that there are tensions within the system of government. On the one hand, the states want to do their own thing, on the other hand, some federal policymakers are sort of trying to push up a federal common standard. Similarly, at the state level, on the one hand, there are a bunch of cities around the country that would like to do their own thing.

And in some of the states, state level policymakers are pushing back against those efforts. And so this raises the natural question, where should authority establish a minimum wage lie? And of course, different academic disciplines are gonna have different takes on this question. So you'll get different answers if you speak with a professor of law, a professor of political science, or a professor of economics.

In economics, it's this toolkit that we call fiscal federalism, which provides us with a coherent framework for thinking about efficient policy making. In this sense of who should have authority or who might best have authority for making policy in a particular space. Second, policy space, Medicaid or publicly financed health insurance for low income individuals.

Some facts about state Medicaid programs. So, first, Medicaid is jointly financed by the federal government and by state governments. Second, in 2010, with the passage of the Affordable Care Act, the federal government sought to induce states to expand their Medicaid programs by threatening to take away a substantial amount of federal funding.

In particular, the form of the threat was if you do not expand your Medicaid program in compliance with the particular rules laid out in the Affordable Care Act, all of the funding associated with your Medicaid program would have been withdrawn. Third, the Supreme Court, although it upheld the Affordable Care Act as a whole, deemed this particular provision to be unduly coercive in a sense that I have no expertise to adjudicate as an economist.

But that was applying sort of previous readings of constitutional law as relates to the relationship between the state governments and the federal government. And that was in the case NFIB versus Sebelius. And as a result of that, the generosity of states programs diverged in a pretty significant way in coming years, in a way that hadn't been intended by the Congress.

And so again, we have this question, where should authority to design this particular policy, in this case, public health insurance for the poor lie. And again, the law, political science and economics are gonna have their own ways of answering this question. And fiscal federalism is going to be the toolkit that economists would apply for thinking about good policy making in this important economic space.

Okay, third and last is a kind of motivating example, school finances. So schools are, of course, primarily controlled at the local level by school districts. About half of their financing comes at the local level. Second, substantial funding for school-districts also comes from state governments. So again, very close to half, or about 45%, of the funding for school districts comes from state governments.

Third, the federal government also kicks in some funding for school districts, along with some rules that are attached to kind of qualifying for receipt of those funds. And fourth, state courts have historically deemed the finances of many states low-resource schools to fail, what's called an adequacy standard, as embedded in the state constitution, not the federal constitution.

And that's led to requirements that resources be either redistributed or at least increased in those relatively low-resource schools. And so, once again, we have the question, where should the authority for financing school districts lie? Once again, different academic disciplines will have different ways of thinking about this problem.

And fiscal federalism is going to provide some guiding principles for economists as we consider this third important policy space. Okay, so now let's get into the public finance perspective on federalism. Modern public finance has a framework for analyzing how federal systems can usefully be arranged from an economic perspective, and that framework is the theory of fiscal federalism.

Now, of course, in this hour, we can only scratch the surface on the details of the theory of fiscal federalism. But what I'm gonna do in the next four slides is sort of quickly introduce you to some of the core forces or core considerations that sort of are key parts of that theory.

Some of which are going to push in the direction of highlighting when decentralization would be preferred, ie when responsibilities will most effectively be carried out at lower levels of government. And some of which are gonna highlight the sets of circumstances or the types of policy spaces in which centralization might be preferred.

And we'll start with the former, and then move to the latter. So a first rationale for decentralization is just going to reflect the simple fact that people differ in their preferences. So the general insight of Wallace Oates in his decentralization theorem is quite straightforward. So it's simply that if people differ in their preferences for public services, do we wanna spend a lot on education or policing, or a little?

Do we want conservative social policy or progressive social policy? Then it would, of course, be inefficient for all of them to receive a uniform package of both social and economic policy from the federal government. And by extension, this implies that there would be potential gains from letting local governments tailor benefits.

So again, this first consideration is quite straightforward. Wallace Oates famously wrote the book called Fiscal Federalism, and so he comes first, he gets top billing and talking about some of these factors. But additionally, if you think about policy debates as they kind of unfold in major newspapers and in academic discourse, there's a lot of pressure that pushes in the direction of federalizing or centralizing policy and making things more uniform.

And so it is important to keep in mind this basic fact that if people differ in their preferences, that the kind of centralized or fully standardized approach may not be the optimal approach. Second rationale for decentralization, which gets more nuanced and in some ways much more compelling from a purely economic perspective, is what we call the Tiebout model.

Which highlights the important role for competition across local governments, which, of course, the federal government does not face by the very fact that it is the federal government. So the Tiebout model is truly profound and that it highlights that even if you make economists sort of favorite assumptions about both politicians and about taxpayers, ie, that they're self motivated and hence aren't thinking directly about the common good.

But are trying to, taxpayers are trying to get social services without paying taxes, and politicians are running around looking for opportunities to potentially skim a little bit off the top or sort of push things in their preferred direction. So what Tiebout showed in his classic 1956 paper, is that competition across governments creates pressures that these sort of standard human motivations notwithstanding, pushing the direction of efficient delivery of public services when it is done at a relatively local level.

So that different localities have to compete against one another. And the logic here is the same standard logic for why prices get driven to marginal cost in standard consumer markets. Which is to say, if I'm considering a move to either Chicago or New York, and I see that the politicians in Chicago are skimming an extra 2% off the top relative to the politicians in New York.

But they're providing the same overall package of public services, then I'll vote with my feet and I'll go to New York. And so, to the extent that the populace is mobile and is actually putting this pressure on politicians to deliver public services efficiently, we can see relatively efficient outcomes again, if we have decentralized, so that the local governments are actually competing against one another.

If we centralize that force goes away. Tiebout also analyzed, consistent with the forces from Oates decentralization theorem, that doing things at the local level, of course, also allows for kind of differences in preferences for public services to be expressed. That also comes with some competitive forces, ie, local governments kind of tailoring benefit packages to attract particular types of people, who have particular wants in terms of what's provided by their governments and so that force remains relevant in the Tiebout model as well.

Okay, what about pushing in the other direction? So what are the forces that might give rise to centralization as opposed to decentralization? So a first insight involves the fact that the benefits of some public goods are dispersed in a way that means, that they flow to individuals that go outside of relatively local jurisdictions.

What's the general insight here? Well, the general insight is that economic efficiency, in the sense typically analyzed in the fiscal federalism literature, requires that boundaries be drawn so that taxes are drawn from precisely the set of residents who benefit from the relevant public goods. Now, of course, this isn't an argument for centralization per say, so to the extent that the benefits of local policing services, fire protection services and the like are indeed very concentrated locally, this would highlight that those services should indeed be provided at the relatively local level.

But to the extent that you have sort of truly national or even global public goods, like national defense, or say, like knowledge production, generating, new technologies, new innovations, to the extent that the benefits of those services are not captured at any particularly local level. What the theory of fiscal federalism highlights is that there's a rationale and a reason for those services to be provided at a relatively central level.

And in Mancur Olson's 1969 classic on the principle of fiscal equivalence, he highlighted or sort of drew out this logic a bit further to point out that given the different scopes of the benefits associated with, say, police and fire protection versus road and transportation networks versus national defense versus utilities.

You could actually see a rationale for a system that might not look so different from the US system of fiscal federalism. Which, to be clear, is not to say that the US system is operating at some hyper efficient level, but is to say that you can plausibly rationalize the complex web of government entities that we have in the US, by applying this powerful line of logic.

Finally, there are factors that are related to social desire for redistribution, the topic of the previous talk by Josh, which also tend to involve rationales for having activities operated by the federal government. The general insight here is that local governments might be ill equipped to engage in redistribution, precisely because they are small and because their populations can flee to other jurisdictions.

And you can think about this as potentially operating with respect to both inflows of people seeking to take advantage of the benefits and outflows of the people who are being taxed, in order to finance those benefits. So if a small state that's located near a large metropolitan area like Delaware, with its proximity to New York, and Washington DC was operating an extremely generous welfare program relative to all its neighbors.

You could imagine that its finances might eventually be overwhelmed by in migration of low income individuals, making the rational decision to try to take advantage of those benefits. And similarly, if that small state was attempting to enact a highly progressive tax on its high income residents in order to finance that benefit, you might imagine that they might say, I can get a much lower tax rate if I just move across state lines over to Pennsylvania and you might see the tax base evaporate under Delaware's feet.

And so this is gonna create a force that's gonna highlight that to the extent that we're kind of taxing and redistributing, that may tend to be sort of better operated at the federal level of government than at relatively local levels of government. So, to summarize some broad implications of these sort of major pillars within the theory of fiscal federalism.

First, services for which benefits are locally concentrated will tend to be best financed and delivered by the state or local governments. Second, redistribution and the provision of goods like national defense will tend to be best financed and delivered by the federal government. Third, on the revenue raising side of the equation, progressive taxes like the income tax, will tend to be more efficiently implemented by higher levels of government.

And fourth, getting a little more subtly into the interactions between layers of government, if key features of redistributive programs are best designed locally, it might be ideal to generate the required revenues federally and then to distribute those revenues to the state or local governments for the implementation phase.

And that's where intergovernmental grants come into the picture. So how do things look in practice? Well, as kind of highlighted when I was talking about the third kind of pillar of fiscal federalism and Carolsen's insight, there's a key sense in which the complex web of governments in the US, plausibly aligns reasonably well with some of these basic or core insights.

So, first, the federal government is indeed the entity that finances and provides both national defense and social insurance for the elderly. Second, the state and local governments are indeed the entities that tend to deliver police and fire protection, education, utilities, parks, things of this nature. Third, the federal government provides some of the financing for interstate public goods like highway infrastructure.

Fourth, the federal government also provides partial financing for redistributed programs like. Health insurance for low income individuals through Medicaid, cash welfare assistance, and recession specific supplements to unemployment benefits. And fifth, the federal government relies on individual income taxes. That is the most progressive of the major tax bases to a much greater extent than do the state or local governments.

Okay, the remainder of the talk is largely going to be now a dive through data to get a sense of how things have kind of evolved over time in the US, first in terms of the kind of aggregate roles of the local, state and federal government. And then in terms of breakdowns in terms of the major activities in which they're engaging.

So as we look at things over time, expressed as a share of GDP, what we see in the graph that you're looking at now is a few kind of broad and important points. So first, the federal government has always been the largest of the layers of government, but its share of overall government activity has tended to shrink over time, rising during crises.

So rising massively during World War II, moderately during the Korean War, moderately during the Great Recession, and again pretty substantially in the context of the recent COVID-19 pandemic. Second, it's interesting to note that local governments have actually been pretty stable in their overall size, expanding a bit as a share of GDP, but having been quite substantial, even if one looks all the way back to the 1930s, to the Great Depression period.

And then third, we see that over time, the place where growth has really occurred has been in the, this is expenditures, but you would also see on the financing side of the ledger that the state governments have really, really grown. So they were between 3 and 4% of GDP of economic activity if we go back to mid century, to the 1950s, and now are accounting for roughly 10% of GDP.

And are actually, I believe, in these last couple of years, in part because of the way that financing for the COVID-19 pandemic expenditures has been handled will actually exceed local government in terms of their total expenditures for the first time. So that's how things have kind of been evolving broadly.

Now to some extent, as we'll see when we look at some of the details, you'll see that the growth of state governments is driven in meaningful part by the fact that they are the primary finance or they are the kind of the entity that operates the Medicaid program, health insurance for low income individuals.

And that program has expanded substantially over this time period. And so that's where a substantial amount of that growth is going to show up. But in general, you can see sort of a moderate increase in decentralization over this time period in the sense that the federal government's share of total government activity has kind of moderately declined as it's remained relatively stable over the last half century, and state governments have substantially expanded.

Okay, next we're going to look at sort of three snapshots of the federal spending pie. And there's kind of a long running way of sort of equipped to describe the federal government, which is that it's a combination of an army and an insurance company. If you go back to 1952, during the Korean War, before the Medicare and Medicaid programs had been legislated, which occurred in 1965, at that time, the federal government is basically just an army.

To go to 1972, Social Security is now sort of aging into something that's starting to approach substantial expenditures. Medicare and Medicaid both existed. And so we see that the healthcare and the pension component of federal spending have risen substantially as a share of the total pie. Defense still remains substantial.

You could think of this as the period where you could describe the federal government as being an army that also happens to operate an insurance company on the side. And now we get to the last year prior to the pandemic. We see that the healthcare piece of the pie and the pensions piece of the pie have expanded pretty massively.

The army is still substantial, but has declined. And so now this is the phase where we think of the federal government as basically being an insurance company that happens to have a pretty substantial and impressive standing military. Okay, let's look at the states. States in some ways are sort of less interesting in the sense that the things that they do have not sort of changed as substantively as this shift away from defense and towards the retirement and health benefits on the federal side.

So states, sort of for this full 70 year period over which I'm showing you data, will have been substantially involved in distributing welfare benefits, in financing transportation infrastructure, and in delivering some mix of education and healthcare services. But we will see, as we come to 2019, that the healthcare piece and the pension piece for the state's own workers have expanded substantially over this time period.

Education has remained substantial. Transportation and cash welfare assistance have declined substantially as a share of the total pie. Part of what's going on here is that redistribution today is being done to a much greater degree through the provision of health insurance benefits than through cash assistance. And so one of the key phenomena here is a decline in the cash welfare slice of the pie and an increase in the healthcare slice of the pie.

Turning finally to the local governments, we'll see that local governments have been involved in a pretty common set of services consistently over this 70 year period. Key pieces of the pie coming from education, transportation, and protection services. Again, that's fire and police protection. We dial it up to 2019.

We continue to see substantial piece of the pie involving education services, fire and police protection, and transportation services. So state and local governments, both in terms of their share of total economic activity and how that's distributed across activities, has been pretty stable over the long haul.

Okay, another piece in which I've taken particular interest in the context of my own research is the role of the federal government in generating financing that's then used to provide funds to state and local governments.

And also to try to kind of cajole state and local governments to do particular things that the federal government wants them to do. What we can see in this graph is that these intergovernmental grants, that is, the revenues that are sent from the federal government to the state and local governments, have grown pretty substantially over the last 70 years.

So if we go back to the 1950s, these grants accounted for less than 1% of GDP. We look at the bulk of the 2010s, they're starting to approach 4% of GDP. And if we look down to during the pandemic, they've escalated to reach as high as five and a half percent of GDP.

So substantial amount of economic resources involve money that's either generated or debt financed by the federal government and then sent along to the state and local governments. And what functions has that been serving? Well, if you go back to the middle of the 20th century, what you'll see is that the primary role of intergovernmental grants was to finance cash welfare assistance and the highway network.

So you think Eisenhower administration and the building of the American highway system along with cash welfare assistance. As I've mentioned, cash welfare assistance or income security has kind of declined in importance over time. And what's really expanded in importance over time has been the Medicaid program, health insurance for the poor, which again, is jointly financed by the state and federal governments.

And you can see that that federal contribution to state Medicaid programs has pushed up into the order of two and a half percent of GDP in recent years. So accounting for a pretty substantial share of these intergovernmental grants, which in their own right are now in these pandemic years, accounting for 1 in $20 sloshing around in the US economy.

Okay, so what are the broad trends here? Over the last 70 years, local governments have been quite consistent in their focus on education, transportation and protection services. Second, the role of states has expanded substantially over this time period, and again, this largely reflects their role in financing health insurance for the poor, which has become substantially more expensive over time.

And that's worked along two axes. One is that the Medicare program has become more generous in its eligibility rules, so that a much larger fraction of the population is eligible for that free coverage. And the second is that healthcare for the poor has become more expensive, just like healthcare for the elderly or healthcare for middle to high income individuals.

And so the cost of each beneficiary has also escalated substantially over time. Third, the federal government's core functions have not changed super meaningfully. Although on the health and retirement side, the Medicare program, of course, did not exist in 1950, it was enacted in 1965, so that piece is certainly new.

But its role as a provider of military defense and of social insurance in the form of retirement benefits and health benefits has been kind of constant in some sense. The composition, however, changing radically, with the spending on defense declining as a share of the federal pie and the spending on health and retirement benefits expanding substantially.

And fourth, there's a growing tendency for the federal government to exert influence over the spending of state and local governments with this coming in large part through the incentives that are embedded in some of these intergovernmental grants. But has also come in the form of mandates and other provisions that are a bit more difficult to track in terms of what their actual economic bite might be from year-to-year.

And this highlights an important sense in which, although when we looked at the evolution of state, local and federal spending over time, it looks like the state share has increased quite substantially, and indeed it has. But in terms of thinking about where the nexus of control truly lies, it's a little difficult to parse because much of that expenditure is incentivized by the federal government or controlled through mandates that are enacted by the federal government.

Okay, in terms of data, let's close out by taking a look at the taxation side of the picture, starting with federal taxes snapshots from 1952 and 2009. There are two key takeaways here. So first, as we'll see, the federal government, as suggested by some of the kind of core tenets or pillars of fiscal federalism.

Will indeed emerge as the government entity that relies to a significant degree on the relatively progressive individual income tax base. So when we look at federal revenues in 1952, we see the progressive income tax accounting for 75% of federal revenues. When we bring it forward to 2019, we see that that share has declined to 56%, but still accounting for a majority of those revenues.

And where does that decline come from? Well, it's come in large part from the fact that the payroll taxes that are used to finance a portion of the Medicare program and all of the Social Security program have increased substantially because the cost of those programs have also escalated substantially.

And so we see the social insurance contributions rising quite substantially as a piece of the federal revenue raising picture. When we look at the states, we see that large chunks of revenue flow to the states in the form of ad valorem taxes, so you think of those in the state context as primarily being sales taxes.

That category also includes property taxes, which is gonna show up in a big way when we look at the local governments. We also see that there are social insurance taxes, there are some income taxes, and there are also a variety of miscellaneous charges and fees. Which you can think of as creating sort of relatively direct linkages between the services that individuals consume and the payments that they make to the government.

So think of school tuition, for example, in the college or university context. Sort of an interesting tension here shows up in terms of the overall theory of fiscal federalism, which is that we see that over this 70 year period, there's actually been a non-trivial rise in the use of income taxes by state governments.

And so although the theory of fiscal federalism highlights that there is indeed kind of a pressure that would tend to discourage at least small states from trying to use progressive tax bases. That pressure being that their high income residents might flee the scene and pursue high standard of living in relatively low tax jurisdictions.

It nonetheless turns out to be the case that some states, you can think California, New York, are in fact able to operate kind of non-trivial income taxes. And some of the research from Josh Rao, the previous speaker, has indeed kind of been trying to suss out the extent to which states like California have pushed the envelope too far in terms of creating a scenario.

In which, when they have an opportunity to leave, the relatively wealthy residents of the states might indeed take that up. And a lot of the empirical research that we do in public finance, that kind of fits into this overall frame of fiscal federalism, is kind of trying to sort out how activated are these pressures that root out inefficient service delivery?

Or that kind of prevent state or local governments from engaging in some of these relatively redistributive activities? It's obviously costly to move, so not everyone just gets up and flees the scene when their tax goes up by a dollar more than what was truly necessary to deliver public services.

But figuring out kind of when states have gone too far, when they push the envelope too far, or simply become too corrupt or inefficient to keep their tax-base around is one of the things that we actively research in this area. And in the context of the shakeup from the Covid-19 pandemic, I would speculate that this will be a very interesting area of research in the coming years, it remains fresh.

Okay, finally, looking at the local tax picture, we see that local governments are sort of even less reliant on these relatively progressive tax basis. So the income tax, it exists, New York City has its own income tax, but in the national picture, it's just a sliver in the pie.

The majority of local government revenues come from ad valorem taxes, in the local context, that's primarily the property taxes that are used primarily to finance school districts. And local governments are also doing a substantial amount of raising revenue through fees, you could think of tolls for using roads, fees collected for using public street parking, things of that nature.

And again, the local governments have really been stable relative to the state governments or the federal government in terms of how things have evolved over the last 70 years. Okay, so that's it for the kind of meat of the facts in the presentation, I'm now just gonna quickly summarize, and then I'll turn it over for a good 30 or 25 minutes of Q and A.

So just by way of closing thoughts, you should think of fiscal federalism as being a framework. In particular, it's the framework that's applied in the discipline of economics for analyzing a bunch of recurring debates that involve the design of federal systems. And the assignment of responsibilities across layers of government within federal systems.

You read the New York Times, sort of other policy outlets. Proposals that we see in the political space often come in the form of asking the federal government to do more, to either centralize or standardize or take on new functions. Public finance as a field, within economics as a whole, teaches us to stop and ask two questions.

The first is, is there a rationale for government to undertake that activity at all? That's the kind of domain of public finance writ large, the second is, if so, is the federal government actually the most sensible entity to take on the particular function in question? And that's what we've been talking about here in the context of my lecture, which is the theory of fiscal federalism.

Which is the toolkit that public finance brings to the table for trying to answer this important question about the design of the US federal systems and federal systems around the globe. So I will close out there, and I'm happy to open it up for Q and A for the remainder of the session.