The Good Society
November 2017

In the world’s rich democratic nations, government taxes and spends a significant portion — 30% to 60% — of economic output. This isn’t, for the most part, a consequence of rent-seeking special interests or narrow-minded bureaucrats expanding their turf. It’s a product of affluence. As people and nations get richer, they are willing to allocate more money to insure against economic risks such as income loss and unexpected large expenses. They also are willing to pay more for fairness — extension of opportunity and security to the less fortunate.¹

But as government gets bigger, does the economy suffer? It’s easy to see how it could. Bigger governments may be more prone to adopt policies that stifle business, reduce competition among firms, or waste resources. They may run up debts that channel resources into interest payments instead of productive activity. High taxes may weaken financial incentives for innovation, investment, and work effort.

On the other hand, government can boost the economy in various ways.² When government protects safety and property and enforces contracts, it facilitates business activity. Enforcement of antitrust rules enhances competition. Schools boost human capital. Roads, bridges, and other infrastructure grease the wheels of business activity. Limited liability and bankruptcy provisions encourage risk taking. Affordable high-quality childcare increases parental employment and boosts the capabilities of less-advantaged kids. Access to medical care improves health and reduces anxiety. Child labor restrictions, antidiscrimination laws, minimum wages, job safety regulations, consumer safety protections, unemployment insurance, and a host of other policies help ensure social peace.

An economy will be healthier with some government than with none, but there surely is some point beyond which government spending and taxing hurts the economy. The question is: Where is that tipping point?

It would be nice if social scientists could locate the tipping point with a theoretical model or a computer simulation. Alas, we can’t. We need empirical evidence. There are many potential sources of such evidence: observational studies of individual behavior, experiments, comparisons across cities or regions, and more. The most informative evidence comes from the experiences of the affluent nations. This is a small and diverse set of countries, and they haven’t been randomly assigned to groups in an experimental design, so we need to be cautious in drawing inferences. But there is no substitute for trying to learn what we can from these countries’ experiences. What do they tell us?

Skip to:

• Does big government mean a heavy regulatory burden on business?
• Does big government mean bad government?
• Do big governments run up large debts?
• Does big government reduce competition?
• Is big government bad for innovation?
• Is big government bad for employment?
• Is big government bad for economic growth?
• Other evidence?
• Summary

DOES BIG GOVERNMENT MEAN A HEAVY REGULATORY BURDEN ON BUSINESS?

Public provision of social services and transfers is paternalistic. Government takes money from us and spends it to ensure economic security, expand opportunity, and enhance living standards. In doing so it reduces individual freedom.³

That isn’t especially objectionable. A few diehard libertarians may believe that individual liberty should trump all other considerations, but virtually everyone supports government paternalism in the form of property protection, traffic lights, and food safety regulations, to mention just a few examples. And many people support public social programs. When basic needs are met, we tend to prefer more security, broader opportunity, and confidence that living standards will improve over time. We are willing to allocate some of our present and future income to guarantee these things, and we are willing to allow government to take on that task. That’s why public social programs tend to expand in size and scope as nations grow richer.

When countries opt for a large public safety net, does government also tend to impose heavy regulations that make it difficult for firms to operate effectively? The World Bank regularly assesses the ease of doing business in most of the world’s countries. Each country is scored on how easy it is to start a business, deal with construction permits, get electricity, register property, pay taxes, trade across borders, get credit, protect minority investors, enforce contracts, resolve insolvency, and flexibly employ labor. The scores on these eleven aspects of doing business are then combined to establish an overall score for each nation. These scores are shown on the vertical axis in figure 1.

Some countries with high government spending, most notably Italy, Belgium, and France, do score low on ease of doing business. But others, such as Denmark, Finland, and Sweden, score quite high. These latter nations tax and spend heavily while still giving economic actors considerable freedom to start and operate a business, allocate capital, hire and fire employees, and engage in all manner of economic activities. This approach is sometimes called “flexicurity.” Government permits individuals and firms substantial economic liberty, with one exception: they pay a significant share of their earnings to the collectivity. This revenue is used to enhance security and opportunity and to ensure that prosperity is widely shared across the society.

DOES BIG GOVERNMENT MEAN BAD GOVERNMENT?

The more a government taxes and spends, according to some, the more it invites lobbying by interest group for favors and the more opportunity and incentive it creates for policy makers and other public officials to dispense such favors.⁴ Larger governments may also create more layers of bureaucracy, impeding effective decision making. Does government performance worsen as its revenues and expenditures increase?

There are various ways to measure the quality of government.⁵ A common measure is the World Bank’s “government effectiveness” indicator, which attempts to gauge public and expert perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies. Figure 2 shows the relationship between countries’ scores on this measure and their level of government spending. There is no association. Countries with bigger governments don’t tend to have less effective ones.⁶

Another potential problem with government policy is excessive complexity — complicated government programs with an array of overlapping rules, benefits, and exemptions.⁷ The United States, for instance, has an assortment of programs and regulations that facilitate access to medical care: Medicare, Medicaid, S-CHIP, the Veteran’s Administration, tax breaks for employer contributions to employee health insurance, health care exchanges run by the federal and state governments in which private insurers compete for customers, a requirement that private insurance plans don’t exclude people with preexisting conditions, a requirement that private plans allow parents to include their children up through age 25, and much more. Our tax system, with its multitude of deductions and exemptions, is equally complex. The IRS Taxpayer Advocate Service estimates that the direct and indirect costs of complying with the US tax code total more than $150 billion a year, or 1% of GDP.⁸ The chief beneficiaries are industries, firms, and affluent individuals who lobby for and are best positioned to take advantage of the multitude of specific provisions and exemptions. Simpler would be better.

Policy complexity in the United States is mainly a result of government’s structure rather than its size. The policy making process is ridden with veto points that allow legislative opponents and interest groups to insert loopholes and special benefits in exchange for allowing proposed policies to go forward. The fact that we have multiple layers of government — federal, state, local — adds an additional layer of complexity.

But don’t larger government programs inherently tend to be more complex? No. Social Security is one of our biggest government programs, but it also is very simple. A “Medicare for All” healthcare system would increase government expenditures’ share of GDP, but it would be much less complex. A tax system with fewer loopholes could raise more revenue and be simpler.

DO BIG GOVERNMENTS RUN UP LARGE DEBTS?

“To spend is to tax,” Milton Friedman once averred. But what if governments that spend a lot don’t, or can’t, generate enough revenues to pay for their expenditures? They may accumulate large debts, the repayment of which diverts resources away from productive economic activity.

As it turns out, that’s not what we observe when we look across the world’s affluent democratic nations. As figure 3 shows, there is no correlation between government expenditures and government debt. Indeed, some of the countries that spend the largest portion of their GDP, the Nordic countries, are among the most likely to raise enough revenues to pay for those expenditures.

DOES BIG GOVERNMENT REDUCE COMPETITION?

Competition drives innovation and economic dynamism. Does an increase in the size of government tend to weaken competition? On one view, the answer is yes. According to Luigi Zingales, “When government is small and relatively weak, the most effective way to make money is to start a successful private-sector business. But the larger the size and scope of government spending, the easier it is to make money by diverting public resources. After all, starting a business is difficult and involves a lot of risk. Getting a government favor or contract is easier, at least if you have connections, and is a much safer bet.”⁹

What do we observe when we compare across rich nations? The vertical axis of figure 4 has an indicator of the degree of competition in product markets. Competition is measured here as the degree of intensity of local competition, the degree to which corporate activity is spread across many firms rather than dominated by a few, the degree to which anti-monopoly policy effectively promotes competition, and the absence of barriers to imports. The scoring for each of these elements is based on a survey of executives conducted by the World Economic Forum. Though not a foolproof measure of competition, this is likely to be a reasonably accurate one.

The figure shows that nations with big governments are just as likely as those with small governments to have competitive product markets.

Why is it that bigger governments don’t necessarily reduce competition? One reason is that firms and other economic actors may care more about shaping government regulations in their favor than about getting government money. Consider the United States. Government taxes and spends less in the US than in most other rich countries, and Americans embrace competition. Yet the US economy is riddled with rules, regulations, and practices that inhibit competition or privilege particular firms and industries. Half-hearted antitrust enforcement allows corporate behemoths to maintain market share and profitability despite little innovation. Patents limit competition in pharmaceuticals, computer software, entertainment, and a slew of other product markets.¹⁰ Licensing, credentialing, and certification requirements for occupations or particular types of businesses dampen competition in product markets ranging from medical care to legal services to education to taxi transportation to hairdressing and beyond.¹¹ Zoning restrictions and historic preservation designations limit expansion of housing units in large cities by imposing building height restrictions and preventing new construction on much of the land.¹² The federal government’s practice of allowing some banks to become “too big to fail” gives those banks an advantage over competitors; they can engage in riskier strategies, with potentially higher profit margins, than can others.¹³

A second reason is that firms’ efforts to get money from government include lobbying for preferential tax treatment. When they succeed, the result is less government revenue, not more — a smaller government rather than a larger one.

Third, the hypothesis that big government results in less competition fails to consider the types of programs on which government spends money. Public insurance programs mainly transfer money to individuals; they offer little opportunity for firms or interest groups to grab a piece of the pie. That is largely true of government provision of services as well. Opportunity for large-scale diversion of public resources is present mainly in government service programs that rely on private provision, such as the US military or Medicare’s prescription drug benefit.

The hypothesis that higher government spending will lessen competition in product markets is a plausible one. But across the world’s rich countries, that’s not what we observe.

IS BIG GOVERNMENT BAD FOR INNOVATION?

It’s common to assert that higher taxes and government spending reduce innovation. Daron Acemoglu, James Robinson, and Thierry Verdier offer an interesting version of the argument.¹⁴ They hypothesize that countries choose between two types of capitalism. “Cutthroat” capitalism provides large financial rewards to successful entrepreneurship. This yields high income inequality, but it stimulates lots of entrepreneurial effort and hence is conducive to innovation. “Cuddly” capitalism features less financial payoff to entrepreneurs and more generous cushions against risk. This yields modest income inequality but less innovation.

Acemoglu and colleagues say their model might help us understand patterns of innovation and economic growth in the United States and the Nordic countries — Denmark, Finland, Norway, and Sweden. America’s cutthroat capitalism successfully fosters lots of innovation. Because of technological spillover from the US, the Nordics, which have opted for cuddly capitalism, enjoy economic growth comparable to the US. But if America were to decide to go cuddly, innovation would slow, according to the model, and both sets of nations would then grow less rapidly.

Are incentives to innovate really weaker in the Nordic countries? Consider Sweden. It has less income inequality and more generous public social programs than the US, but while Swedish CEOs and financial players don’t pull in American-style paychecks and bonuses in the tens of millions, there is little to prevent a Swedish entrepreneur from accumulating large sums. In the 1990s Sweden instituted a major tax reform, reducing marginal rates and eliminating loopholes and deductions. Corporate income and capital gains tax rates were lowered to 30%, and the personal income tax rate to 50%. Later the wealth tax was done away with. In the early 2000s, a writer for Forbes magazine mused that Sweden had transformed itself from a “bloated welfare state” into a “people’s republic of entrepreneurs.”¹⁵

If incentives to innovate are significantly weaker in Sweden than in the US, when did the difference emerge? An indicator of financial incentives for entrepreneurs is the top 1%’s share of household income. An indicator of the extent of cushions against risk is government expenditures’ share of GDP. Both are shown in figure 5, going back to 1910. The two countries were similar until the second half of the twentieth century. Government spending began to diverge in the 1960s, and income inequality diverged in the 1970s.

According to Acemoglu and colleagues’ logic, then, incentives for innovation in the US were weakest in the 1960s and 1970s. Yet there was plenty of innovation in the 1960s and 1970s, including notable advances in computers, medical technology, and other fields.

Given that the (hypothesized) cross-country difference in incentives emerged in the 1960s or 1970s, if modest inequality and generous cushions are bad for innovation we would expect the Nordic nations today to be far behind the United States. However, innovation rankings consistently place the Nordic countries on par with the US, or only slightly behind. The World Economic Forum’s Global Competitiveness Index for 2015-16 rates Finland as the world’s second-most innovative nation, ahead of the United States, with Sweden and Denmark close behind. The 2015 Cornell-Insead-WIPO Global Innovation Index ranks Sweden third, the United States fifth, Finland sixth, and Denmark tenth.¹⁶

If Acemoglu and colleagues are correct about the value of financial incentives in spurring innovation, we should see this reflected not only in the United States but also in other nations that have low-to-moderate government spending and relatively high income inequality, such as Australia, Canada, Ireland, New Zealand, and the United Kingdom. But as figure 6 indicates, the innovation rankings suggest that these other cutthroat capitalist countries tend to be less innovative than the Nordics, not more.

The comparative and historical data offer little, if any, support for the notion that big government is bad for innovation.

IS BIG GOVERNMENT BAD FOR EMPLOYMENT?

Working-age French, Belgians, and Germans spend, on average, about 1,000 hours a year in paid employment. In the United States, Switzerland, and Japan, by contrast, the average is 1,200 to 1,300. That’s a big difference. Is it due to differences in the size of government?

These averages are determined by the share of people who have a paying job and by the number of hours they work over the course of a year. In the United States, for instance, the employment rate in 2014 was 68% and those employed worked an average of 1,790 hours (.68 x 1,790 ≈ 1,200). In France, the employment rate was 64% and the average number of hours worked was 1,490 (.64 x 1,490 ≈ 950).¹⁷

Because they reduce the financial reward to paid work, high taxes may reduce employment.¹⁸ On the other hand, some people might work more when taxes are higher, in order to reach their desired after-tax income. And lots of other things affect people’s calculations about whether and how much to work, including wage levels, employment and working time regulations, paid vacation time and holidays, availability and generosity of government income transfers, access to health insurance and retirement benefits, the cost of services such as childcare, and preferences for work versus leisure.¹⁹

Figure 7 shows the association between government size and employment across the rich countries. The pattern is consistent with the notion that high taxes reduce work hours.

But knowledgeable comparativists will notice a familiar clustering of countries in this graph.²⁰ One group, in the lower-right corner, includes France, Belgium, Italy, Germany, and the Netherlands. These countries, along with Austria, have several features that might contribute to low employment hours. One is strong unions. Organized labor has been the principal force pushing for a shorter work week, more holiday and vacation time, and earlier retirement. These nations also have been characterized by a preference for traditional family roles — breadwinner husband, homemaker wife. This preference, often associated with Catholicism and Christian Democratic political parties, is likely to influence women’s employment rates and work hours. It is manifested in lengthy paid maternity leaves, lack of government support for childcare, income tax structures that discourage second earners within households, and practices such as German school days ending at lunch time and French schools being closed on Wednesday afternoons. These countries also fund their social insurance programs via heavy payroll taxes, the kind most likely to discourage employment growth in low-end services.²¹

A second group consists of the four Nordic nations: Denmark, Sweden, Finland, and Norway. These countries too have strong unions. But they also have had electorally successful social democratic parties, which have promoted high employment.²² Denmark and Sweden, in particular, have led the way in using active labor market programs to help get young or displaced persons into jobs, public employment to fill gaps in the private labor market, and government support for childcare and preschool to facilitate women’s employment.

A third group of countries, in the upper-left corner, includes Australia, Canada, Japan, Korea, New Zealand, and the United States. These nations have relatively weak labor movements and limited influence of social democratic parties and Catholic traditional-family orientations.

The other five countries — Ireland, Portugal, Spain, Switzerland, and the United Kingdom — are a hodgepodge.²³

Based on their institutional-political makeup, we would expect the weak-labor countries to have comparatively high employment hours, the social-democratic countries to be intermediate, and the traditional-family-roles countries to have low hours. As figure 8 indicates, that’s exactly what we see.

If we adjust statistically for institutional-political group membership, the negative association between tax levels and work hours shown in figure 7 disappears. Differences in union strength, in preferences for traditional family roles, and in the influence of social democratic parties, rather than taxation, are the likely source of differences in employment hours across the world’s rich nations.²⁴

IS BIG GOVERNMENT BAD FOR ECONOMIC GROWTH?

So does big government harm the economy? The best single indicator is economic growth — growth of per capita GDP.

Data on government revenues are available for the United States going back to the early 1900s. The first chart in figure 9 shows that revenues rose from the 1910s through the 1990s and then leveled off. All told, government revenues increased by approximately 25 percentage points, from less than 10% of GDP to around 35%.

The second chart in the figure shows GDP per capita all the way back to 1890. The data are displayed in log form in order to focus on the rate of growth. The straight line represents what the data would look like if the economic growth rate had been perfectly constant. The actual data points hug this line. In other words, despite occasional slowdowns and speedups, the rate of per-capita GDP growth in the United States has been essentially constant for the past 120 years.²⁵ We’ve gone from being a country with relatively small government to one with medium-size government, and in the course of doing so, we’ve suffered no slowdown in economic growth.²⁶

Now let’s look at two big-government countries: Denmark and Sweden. Figures 10 and 11 show trends in government revenues and in economic growth for these two nations. In both, government revenues jumped sharply, especially in the decades after World War II. Revenues stopped rising around 1990, flattening out in Denmark and falling back a bit in Sweden. Just like the United States, these two countries have had a virtually constant rate of economic growth since the late 1800s. A very large increase in the size of government didn’t knock either country off its growth path.

A possible exception is Sweden around 1990. At the end of the 1980s, government revenues in Sweden reached 65% of GDP. Shortly thereafter, the country had a severe economic downturn. By 1995, revenues had dropped to 60% of GDP and the economy returned to its long-run growth path. The onset of the early-1990s crisis stemmed mainly from deregulation of Swedish financial markets and the government’s pursuit of fiscal austerity during the downturn. But given the coincidence in timing, it could be argued that government taxing and spending at 65% of GDP is too high. Maybe that’s correct. If we follow that logic, however, then we must conclude that 60% of GDP, the level of government revenues when the Swedish economy returned to solid growth, is not too high.

When the United States is compared to countries such as Denmark and Sweden, a common objection is that the latter are small and homogenous.²⁷ But the point here isn’t to compare or contrast these countries. The point is that developments over time within each of the three countries tell a similar tale. In each, government taxing and spending rose substantially — in the United States to about 35% of GDP, in Denmark and Sweden as high as 60% — with no apparent impact on economic growth.

Some might still object that only a small, homogenous nation can have a big government without hurting economic growth. The story would be that a large, heterogeneous country such as the US may do just fine with a rise in government up to 35% of GDP, but beyond 35% growth will slow down. It’s conceivable that this is true, but the story is based on assumption rather than evidence, so there is reason for skepticism.

Let’s extend the inquiry to the full set of rich longstanding-democratic nations. Figure 12 shows the average level of government revenues (horizontal axis) and the average economic growth rate (adjusted for starting level, vertical axis) in these countries between 1979 and 2007. There is no association between government size and economic growth. More thorough cross-country studies have tended to reach a similar conclusion.²⁸

Figure 13 looks at changes in government size and economic growth from 1950-73 to 1979-2007. As the horizontal axis indicates, the size of government increased in all rich countries between these two periods, but much more in some than in others.²⁹ Economic growth rates slowed in almost all of these nations, but more in some than in others. If big government reduces economic growth, we would expect to see a larger slowdown in growth rates in countries with larger expansions in government size. But the data once again suggest no such such effect.

OTHER EVIDENCE?

Most of the evidence I’ve looked at here compares across the world’s rich nations in recent years or decades. It would be helpful to consider a longer period and examine changes over time. Unfortunately, with the exception of economic growth, we can’t. In most of these countries government size hasn’t changed much since the mid-1980s, so to assess the effect of changes we need to go farther back in time. But measures for most of the hypothesized consequences of big government — ease of doing business, government effectiveness, competition, innovation, and employment hours — are only available for recent years.

What about comparing across the US states? Here too we have limited data: most of the relevant indicators aren’t available for the states. Also, government taxing and spending at the state level is much smaller in quantity than at the federal level, so it isn’t clear how much state differences in government size are likely to matter for economic outcomes.

SUMMARY

There are lots of reasons to hypothesize that when government spends and taxes more, economic performance may suffer. And there undoubtedly is some level of taxing and spending beyond which it will suffer. The evidence suggests, though, that the world’s rich longstanding-democratic nations probably haven’t passed that tipping point. Countries with higher taxes and government spending don’t tend to have heavier regulatory burdens on business, less effective government, higher government debt, less product market competition, less innovation, less work effort, or slower economic growth.