I have drawn some ire from fellow libertarians for my criticism of their fiscal policy priorities. My review of the book Trumponomics by Steve Moore and Art Laffer, which concluded that tax cuts have become ineffective, rendered a couple of surprising comments from fellow libertarian economists. Nevertheless, as I explained in my series Tax Cuts or Spending Cuts, facts are facts; cutting taxes to close the budget gap in a big welfare state is about as futile as ignoring gravity.
The only way to close the budget gap is by means of structural spending reductions. Since such reforms are no longer being discussed in the public discourse, nor being given attention by leading libertarian thinkers, pundits and scholars, starting today I am rolling out a series of articles on how to structurally reform away the welfare state.
In this the first installment we will condense the case for structural spending reform. I find it necessary to do so, partly – again – in response to the conventional wisdom that tax cuts can save us, but partly also in response to another idea being floated around among conservatives and libertarians: fiscal rule making.
Many people with influence have suggested that all Congress needs in order to end our deficits and prevent a debt crisis, is to follow a set of fiscal rules. A new book from the Cato Institute offers a collection of 20 essays centered around this notion; as I explained in my review of the book, for two reasons fiscal rule-making does not work:
- The rules are “Pippi rules”, i.e., they are self enforced with impunity for non-compliance;
- Without exception, the rules rely on healthy levels of GDP growth in order to work.
The last point applies not only to fiscal rule-making but also to the unending pursuit of tax cuts. In the case of rule-making, the prevailing wisdom is that GDP growth is exogenous to fiscal policy, in other words that it is not affected by the enforcement of a fiscal rule. However, as we will see in a coming article, fiscal rules can actually undermine the very basis for their own enforcement.
In other words, GDP growth is not exogenous, but endogenous to fiscal policy.
Proponents of tax cuts recognize this: in fact, the endogenity of GDP is the very life blood of the tax-cut argument. The problem here, though, is that tax cutters limit the endogenity to taxes; there is no explicit consideration of the effects of government spending on GDP growth. Right there we have the mistake that prevents the tax cutters from seeing how their budget-balancing strategy has run its course and no longer works.
It is essential for any attempt to save us from a debt crisis, that we understand why GDP growth is affected by the size of government spending. Before we get to the analytical explanation, let us start with empirical evidence. Figure 1 summarizes what this evidence says, namely:
- When government spending as share of GDP increases, i.e., when G/Y goes up (horizontal axis), initially the growth rates of taxes (t) and government spending (g) are largely similar;
- As the size of government passes a certain point, the growth rates of t and g divert, with the former declining and the latter increasing;
- when government spending grows faster than tax revenue, we get a structural budget deficit.
Let us review some data that confirm the image in Figure 1. First, Figure 2 reports data from Europe on the relationship between government spending as share of GDP, G/Y, and real GDP. Covering 31 countries, almost all of them over a 25-year period (1995-2019), it paints a stark image of the negative relationship between economic growth and the size of the welfare state:
Next up: numbers from the U.S. economy, which gives us an opportunity to link the size of government to the root cause of the budget deficit: the welfare state.
Government growth in America can be divided into five distinct phases. The first era runs from 1950 to 1963. This is the first phase of stability (Stability 1 in Figure 3 below). During this phase the welfare state was still ideologically conservative and did not engage in economic redistribution. It was, simply, confined to the provision of a safety net for the poor and needy. This was the welfare state that Congress and the Franklin Roosevelt administration created in response to the Great Depression.
The second phase of government growth begins with President Lyndon Johnson’s State of the Union speech in 1964. There, he declares his War on Poverty and marks the beginning of a fundamental overhaul and expansion of the welfare state. New entitlement programs like Medicare and Medicaid are created, others revamped for much more comprehensive purposes.
At the heart of the War on Poverty is a new, relative definition of poverty. A person is no longer poor because he lives below a certain, fixed standard of living. A person is now poor because he earns below a certain percentage of median income. When median income rises – as it does when the economy is doing well – the poverty limit rises as well. More people qualify for government handouts.
As I explained in my book The Rise of Big Government, the significance of this welfare-state metamorphosis cannot be understated.
After this roll-out phase for the new, socialist welfare state, it was time for the implementation phase: government needed to consolidate its new spending programs. Fiscally, this meant that government grew to the new, larger proportions that all the War-on-Poverty legislation prescribed. During this phase, which essentially coincided with the 1970s, government spending increased significantly.
This is also the phase during which government spending outpaced tax revenue on a permanent basis. The structural budget deficit was born.
Once the new welfare state was consolidated, government went into a new phase of relative stability. It lasted from 1980 to 2007, the year before the Great Recession started. During this Stability 2 phase (again, see Figure 3 below), government spending remained largely constant as share of GDP, but that stability was attainable thanks only to two tax reforms. Government spending was still small enough to let tax cuts work – there was a Laffer effect to be counted on.
If it had not been for the Reagan tax cuts fundamentally overhauling the tax code, there would not have been a long, stable growth period through the 1990s; the Bush tax cuts generated a more limited, yet visible growth spurt that helped carry the economy through most of his presidency.
With their growth record, these two tax reforms generated enough growth to essentially keep steady the ratio of government spending to GDP.
It is easy to get the impression from this long phase of stability that America had struck a golden balance between the welfare state and free-market capitalism. That was not the case, as evidenced by the perpetuated deficit. The Reagan and Bush tax reforms were not enough to close the budget gap, and the difference in effect of the two, with the first having stronger effects than the second, remains unrecognized in the literature.
The reason why the two reforms failed to fully fund the welfare state is simply that this structure of government spending grows by its own volition. Government spending is exogenous to economic growth, a fact that originates in the very ideological design of the welfare state.
Phase five, the Stagnation phase in Figure 3, is when the welfare state has grown big enough to permanently depress economic growth. This phase provides ample evidence of how the welfare state overpowers its host economy. During this phase, which begins in 2008 with the Great Recession, total government outlays average more than 37 percent of GDP. Economic growth is so poor that its annual average for the entire period is only 1.7 percent:
Predictably, the decline in growth has taken a toll on government revenue. When taxes have not delivered sufficient money to fund the growing welfare state, government – especially states and municipalities – have resorted to non-tax revenue. As Figure 4 explains, the rise of fees, charges and other revenue sources has coincided in time with the transformation, implementation and growth of the socialist welfare state. However, not even the rise of non-tax revenue has been enough: the deficit reported in Figure 4 (red) is mostly federal but also includes overspending by state governments.
De facto, deficits have become a permanent source of government funding:
It is worth noting that as taxes become less important as a revenue source for government, the effectiveness of tax cuts also declines.
With declining GDP growth under an increasingly burdensome welfare state, government debt keeps rising. Figure 5 compares the debt-to-GDP ratio to real GDP growth. To highlight trends, the numbers are reported as five-year moving averages:
Let us now add together everything we have learned so far about the interaction between government spending, GDP growth and tax revenue, and add one more twist to the tax-or-spending-cuts debate. Figure 6 reports a total of 262 quarterly observations of annual growth rates in all three variables, from 1954 to Q2 of 2019. The observations are not reported chronologically, but are instead organized by GDP growth, from high to low (blue).
Tax revenue correlates positively with GDP growth (black dashed), which is not surprising. What does stand out, however, is the turn upward of government spending growth (red dashed) when GDP growth falls below three percent per year. As GDP growth gets weaker, government spending accelerates:
We have now learned three important things about the government budget:
- As spending grows, GDP growth declines;
- As GDP growth declines, spending growth accelerates; and
- With declining GDP growth, tax revenue slacks off as well.
There is more to be said about the first two points; for now, let us note the conspicuous divergence between GDP and government spending toward the right end of Figure 6. That gap alone explains our structural budget deficit.
However, first, we need to add one last point about taxes. As Figure 6 tells us, tax revenue fluctuates with GDP growth, but the fluctuation is higher in tax revenue. This means, plainly, that tax revenue is a volatile source of revenue. Figure 7 extracts the equations defining the trend lines in Figure 6 and plots them strictly as analytical representations of how these two variables correlate:
The problem with Figure 7 is that the volatility in tax revenue has increased with each supply-side tax reform. This means, in turn, that tax revenue plunges more violently in recessions, but since the long-term growth trajectory for GDP – and therefore the tax base – is lower than it used to be, this volatility is not symmetrical. We don’t get enough of a compensating “Keynesian” surge in tax revenue at the peak of the business cycle.
In other words, we are left with a structural deficit.
But why, then, do we actually have this structural deficit? This question is of course essential to our discussion of structural spending reform. Figure 8 has the answer. It reports the share of the federal budget that is dedicated to entitlement spending, i.e., the welfare state. The blue function represents the welfare state’s share of the budget in 1964, when the War on Poverty began. Today, two thirds of all federal spending goes toward entitlements, from education to Social Security, from the Earned Income Tax Credit to Medicaid and Medicare:
To highlight, Figure 9 divides the 2019 federal budget by major program category. Notice the share that goes to national defense: in 1960 that share was 50 percent.
Spending on entitlement programs grows for reasons that are inherent to the programs themselves. This growth causes a depression in GDP growth, which in turn causes a structural budget deficit. The only way we can close the budget deficit is by reforming away the welfare state.
How? Please proceed to Part 2.
Click the Follow button to always get updates first!
 For a more detailed discussion of the ideological character of the American welfare state, see Larson (2018) and Larson (2020f).
I continue to receive criticism and questions over my articles pointing to the futility of pursuing tax cuts. I am happy to answer all those questions, and I will gladly continue to point to why structural spending reform is the only way to go. I will publish my own reform plan soon after the election; for a glimpse of what it will cover, check out my 2012 book on welfare-state reform.
In the meantime, let me reinforce the point about supply-side economics and further tax cuts. I am not the only one to point to the limitations of what this theory can do; last year Cato Institute fellow Ryan Bourne suggested that taxes have become too low for the Laffer Curve to still be effective. In short, he proposes that the tax cuts that have swept across the industrialized world in the last few decades have taken taxes down to where a cut no longer has the stimulative effect on revenue that it had when Art Laffer first introduced his famous curve. When tax rates were in the “ineffective” segment in Figure 1, a cut would increase revenue; once the rates fall into the “effective” segment, a cut reduces revenue:
Figure 1: The Laffer Curve
Bourne has no data to back up his claim, but in order for his argument to work he would need to show that labor supply no longer responds positively to tax cuts. It does, as shown by every tax cut in the United States the past 40 years. The Trump tax reform was particularly effective in increasing workforce participation.
The problem with the supply-side school is not that taxes are too low for the Laffer effect to work. The problem is that we don’t get enough growth out of the workforce participation that comes from lowering taxes. That problem, in turn, is related to the large presence of government in the economy; there is ample evidence – as I have reported in my book Industrial Poverty – of a 40-percent-of-GDP threshold: once government spending is higher, it permanently depresses GDP growth.
Supply side economists have inadvertently helped government grow. As much as a lot of supply-side libertarians want to dislike me for making this point, the evidence is irrefutable: the tax cuts they have helped bring through Congress were all good and necessary in and of themselves, but since they were not accompanied by any spending-reform efforts of even remotely similar magnitude, all that happened when tax revenue surged was that Congress decided to spend even more money.
If the supply-side school had stood equally strongly on both legs – tax cuts and spending cuts – things would have looked very differently in our economy. As it is now, the tax-cut strategy has allowed government to grow to a point where its size depresses the very growth effect that supply-side economists rely on to make their tax cuts work.
Again, that is not to say the tax cuts weren’t good. They were. But since government has continued to grow, the efforts to cut taxes have at least in part become a shuffle game. The tax burden in the U.S. economy has shifted, and not in a way that we should necessarily celebrate. First, consider Figure 2, which reports the aggregate tax burden on corporate profits:
Figure 2: Effective tax rate on corporate profits
Clearly, U.S. tax policy has moved away from heavily burdening businesses. This is good, of course, even if it for a long time happened through measures that invited advanced – and costly – tax avoidance (which unlike evasion is legal). The federal corporate-tax system we got with the Trump reform was much welcome, but as Figure 3 explains the reduction in taxation of corporate profits has come with a shift toward heavier reliance on other sources:
Figure 3: Composition of U.S. taxes
In other words, individuals bear the brunt of the tax burden. That does not mean it is better to have high taxes on corporations – all taxes should be minimal – but Figure 3 is a hint of what happens when we do good tax reforms without equal emphasis on spending reform.
So long as the balloon is of the same size, a squeeze on it on one size must result in a bubble on the other side.
With taxes shifting over onto personal income and consumer spending, the pertinent question is what this means for households and their finances. Since consumer spending accounts for about 70 percent of GDP, a concentration of taxes onto households will inevitably depress economic activity. The scope of this problem emerges from Figure 4, which reports the striking decline in work-based earnings and the rise in household dependency on government and equity-based income:
Figure 4: Personal income and taxes
We can see a good part of the problem with the Laffer Curve in Figure 4. First and foremost: the three big tax reforms over the past 40 years – Reagan, Bush and Trump – have not lowered the tax burden on personal income. They have helped keep that burden flat, but that is also all they have accomplished.
Furthermore, the composition of private income is such that the intended effects of another tax reform would be limited. Today, wages and salaries account for only half of personal income. This is the share we need to target with tax cuts if we are going to bring about more workforce participation. The Trump tax reform showed, again, that this can be done, but the tax cuts would have to be concentrated to the lowest income segments; the higher a household’s income is, the larger a share of it comes from equity. Cuts in taxes on equity-based income are not as effective in stimulating economic growth.
At the same time, lower-earning households depend to a larger extent on tax-paid entitlements. Their total share of personal income is approximately 17 percent, and will very likely increase further in the future.
The tricky thing with this income source is that households tend to lose entitlements as their work-based income increases. Even if they increase their workforce participation thanks to a tax cut, their total income does not increase accordingly. Thereby, consumer spending is not boosted, depriving the economy of some of the growth the tax reform would otherwise generate.
Again, Figures 3 and 4 tell us with painful clarity that our government has not become cheaper as a result of the practice of supply-side economics. Figure 5 makes this point from a different angle. Suppose we bundled together the cost of federal, state and local government taxes and other revenue sources and created one “all inclusive” personal income tax. What would the rate look like? Figure 5 explores that question with the tax base defined as total personal income (grey), employee compensation (dashed blue) and wages and salaries (solid blue):
Figure 5: The total cost of government as share of personal income
Again, our government is not solely funded through taxes on personal income and consumption, but as Figure 3 showed we are moving in that direction. It is also worth noting that even a corporate income tax is ultimately paid by the employees, the investors and the customers of that corporation. Therefore, the experiment in Figure 5 is quite telling of what the cost of our government actually looks like today.
Bluntly: four decades’ worth of tax reforms have not helped reverse the growth in government. Its cost has tapered off, but we have to keep in mind that Figure 5 does not take budget deficits into account. They really aren’t more than a hidden tax on future personal income.
Long story short: all that matters is structural spending reform. Once that is under way, there will be room for further tax cuts. As a bonus, budget deficits will gradually disappear.
As we have seen so far, our efforts to rein in government spending and balance the federal budget (in that order) have failed. The reason is that people have been focused on the wrong solutions:
- There is too much focus on tax reform; specifically, there is no point in eliminating the income tax, because as state and local governments have demonstrated so carefully, the one tax is always replaced by other revenue sources;
- Supply-side economics has stopped working; the welfare state is simply too big; and
- Once you turn to the spending side, the salami approach – also known as the Penny Plan – will force government to default on its promises in Social Security, Medicare, Medicaid and other entitlements.
So if we want to bring fiscal sanity to Washington, what is the alternative? Structural spending reform.
First, though, one final word on the tax side. The more we reform taxes in order to stimulate economic growth – with the implied goal to increase tax revenue and reduce the budget deficit – the more serious our budget problems get. In addition to the reasons I discussed in earlier installments of this article series, there is also the problem of increased revenue volatility. Every tax reform we have done has increased the amplitude of tax revenue. Consider Figure 1a, which reports the current-price growth rates in GDP (blue) and in government revenue for all levels of government combined (grey):
Figure 1a: Growth rates in taxes and GDP
Specifically, there are three episodes in Figure 1a that illustrate how our ability to balance government finances has been weakened over time:
Figure 1b: Episodes in revenue and GDP growth
Before the Kennedy tax reform, the correlation between tax revenue and GDP growth was fairly close (1). With the Kennedy reform we saw a slight departure of the two, but it was not very pronounced. It indicated, though, what was to come.
In the late 1970s we saw high inflation and very high growth rates in tax revenue (2). The reason, of course, was the combination of inflation indexing in income-tax scales and the high reliance of federal, state and local governments on income taxes. The combination of high taxes and high inflation was bad for the economy, with both of them ending in the early years of the Reagan presidency.
At this time, supply-side economics still worked. Economic growth was solid during the Reagan years, as was growth in tax revenue. The problem was that the federal tax base shifted toward higher incomes, a problem that would become even more pronounced with the Bush tax reform. The swings in tax revenue increased over time (3), making it increasingly difficult for governments at all levels to balance their budgets.
The Trump tax reform appears to have increased the revenue amplitude. This problem is particularly noteworthy since inflation has been low and GDP growth only moderate, removing two factors that otherwise would cause sharp swings in government revenue.
To summarize, the problem with government finances is not the revenue side. It is the spending side. Figure 2 summarizes this point by reporting swings in tax revenue, GDP growth and government spending for the same period of time as in Figures 1a and 1b. The swings are sorted based on GDP growth from high to low. As we move from the origin of Figure 2, outward to the right, and as the GDP growth rate tapers off (blue) so does the growth rate of tax revenue (dashed black), albeit more pronounced.
What is really interesting in Figure 2 is the trend in government spending (dashed red). While remaining relatively unchanged for the better part of the interval of GDP growth rates, it rises noticeably as GDP growth falls below two percent:
Figure 2: Growth rates for GDP, tax revenue and government spending
In short: as GDP growth slows down over time, the discrepancy between tax revenue and government spending increases.
Herein lies the key to government reform. It is not about taxes, and it is not about salami-tactic spending reform.
It is about structural, permanent reductions in government spending. The reason why government outlays increase more prominently when GDP growth tapers off is that the programs that government spends most of its money on are designed to become more important in the lives of Americans precisely for that reason. When GDP grows more slowly, fewer people are able to work their way out of poverty (as statutorily defined) and more people become vulnerable to the thresholds in those programs.
The Earned Income Tax Credit is a classic example, where someone making $35,000 per year can face the same marginal-tax effect from a $5,000 income rise as someone making ten times that amount. More people get locked in to dependency on the welfare state, simply by virtue of how the welfare-state entitlements are designed.
Over time, more people remain dependent on Medicaid, on food stamps and other programs. Even Social Security is affected by slow growth, although that only shows up in its finances over an extended period of time. Still, the effect is worth noting: retirees who earned a relatively modest wage get a larger percent of their income replaced than those who earned more. With low incomes paying modest taxes into the system, over time this means a more rapid depletion of the system – and a relatively higher increase in Social Security spending.
In other words, what we need is a comprehensive, structural reform of all the spending programs that the federal government spends our money on. This kind of reform fundamentally rewrites the role of the federal government by terminating the promises that come with its vast portfolio of entitlement programs. A structural reform gets government out of the business of economic redistribution and strictly bans the use of taxpayer money for this very purpose.
Structural reform means that the private sector resumes responsibility for everything the government does today, except its minimal-state functions: defense, law enforcement and infrastructure.
There are two big challenges in executing structural reform:
- Making sure the private sector can actually take over the responsibilities that government has today without leaving poor and vulnerable citizens worse off; and
- Making the transition from the current system to the new so smooth that it does not disrupt either economic growth or political and social stability.
These are significant challenges, and they can seem daunting. That, however, is no reason to shy away from them.
I have my own plan for this in the making. Until I publish it, here is a first taste of what structural reform can look like.