Get Society Rich Quick: The Ideal Level of Government Spending

It is commonly claimed that government spending is good for the economy, but the statistics show the opposite.  Most recent studies find a negative correlation between total government size and economic growth. Why is economic growth important?  Because wealthier societies are generally happier societies.

No matter how good things get for us, we always want more. Human beings want better:

  • food
  • housing
  • entertainment
  • transportation
  • health services

A certain level of government spending is necessary to maintain an economy that can provide and improve these things. A well-functioning economy requires government to:

  • enforce contracts
  • protect a country from foreign invaders
  • keep people from stealing from each other
  • keep people from hurting each other
Government expenditure on these functions increases a country’s productivity.

Graph of GDP Growth Rate vs Size of Government

Graph Source:

However, when government spending exceeds these mandates, it takes the country’s human and material resource allocation decision-making power away from society’s massive collective of individual intelligences. This is a bad idea because individuals are just better at deciding how allocating resources to satisfy their own desires since:

  • They cumulatively have access to infinitely more detailed and specific information than any small group of politicians and bureaucrats do.
  • Satisfying as many desires as possible (by maximizing productivity) is their sole incentive.

Politicians, on the other hand, do not allocate resources as productively (satisfy fewer human desires) because they:

  • Only have access to general demographic data
  • Are primarily incentivized to obtain campaign contributions to get re-elected

The ability to obtain campaign contributions is primarily dependent on the candidate’s ability to divert the nation’s resources to corporations and special interest groups. Hence, resources are not specifically allocated to maximize their productivity, but are allocated in the pursuit of political goals.

This graph presents data on the relationship between size of government and economic growth for the twenty·three long-standing members of the Organization for Economic Cooperation and Development.

Graph of Total Government Expenditure as a Percent of GDP vs Growth Rate

Graph Source:

The data show that a 10 percent increase in government expenditures as a share of GDP reduces the annual rate of growth by about 1 percent. The data also imply that the size of government in these countries is beyond the range that maximizes economic growth.

The graph contains four dots (observations) for each of the twenty-three countries (one for each of the four decades during the period 1960-1999). Each dot represents a country’s total government spending as a share of GDP at the beginning of the decade and its accompanying growth of real GDP during that decade. As the plotted line in the exhibit shows, there is a clearly observable negative relationship between size of governments and long-term real GDP growth. Countries with higher levels of government spending grew less rapidly. The line drawn through the points indicates that a 10-percentage-point increase in government expenditures as a share of GDP leads to approximately a 1-percentage-point reduction in economic growth.

Graph of GDP Growth In Big and Small Government Countries

The reason GDP matters is that it is directly correlated to a reduction in human suffering. Economic growth is the primary differentiating factor between a life lived in the US and a life lived in North Korea. Economic growth is the primary differentiating factor between a life lived in the 11th century and the 21st century. Economic growth has cured countless diseases which would have otherwise imbued countless lives with unimaginable suffering. Economic growth has prevented billions from dying of starvation. If governments will get out of its way, economic growth will eventually eliminate human suffering from the planet. Also, according to the most recent income/happiness study, the lower a person’s annual income falls below $75,000 a year, the unhappier he or she feels. But no matter how much more than $75,000 people make, they don’t report any greater degree of happiness. The strong correlation between happiness and GDP is illustrated below.

Graph Showing Wealthier Countries Exhibit Greater Life Satisfaction

Graph Source:

Here is a graph from Politics & Prosperity. Look at the historical relationship between government spending and GDP growth:

Historical Graph of US Government Spending and Economic Growth

  (Source notes for this graph and those that follow are at the bottom of this post.)

Historical Graph of Government Spending as a Percent of GDP vs Economic Growth

  Source notes: Estimates of real and nominal GDP, back to 1790, come from the feature “What Was the U.S GDP Then?” at Estimates of government spending (federal, State, and local) come from USgovernmentspending.comStatistical Abstracts of the United States, Colonial Times to 1970Part 2. Series Y 533-566. Federal, State, and Local Government Expenditures, by Function; and the Bureau of Economic Analysis (BEA), Table 3.1. Government Current Receipts and Expenditures, and Table 3.12. Government Social Benefits. Amount spent by governments (federal, State, and local) on national defense and public order and safety by consulting BEA Table 3.17. Selected Government Current and Capital Expenditures by Function.

Please share your thoughts in the comments section!

  • Anonymous

    "If governments will get out of its way, economic growth will eventually eliminate human suffering from the planet."

    China. Socialist. 9% per year. The notion that people are rational actors has been a falling star in economics, and is being demonstrated as nonsense more and more every year in behavioral fields. I find it funny you didn't include a smidgen of this data, especially since it is so obvious to anybody just from their own life experience. I'm sure you've seen people loading up on lottery tickets, plus are familiar with credit default swaps and the like.

    How can you honestly expect that status quo default private market activity will take care of the poor? Unemployment has never fell below 3%, that doesn't even include people who give up looking.

    The real trend has been for the massive accumulation into personal family fortunes. For every argument against minimum wage, there is 10,000,000 real persons who works in those slave like conditions who has nothing to negotiate with. Unchecked growth and capitalism only exist to serve the well being of the shareholders, guess who has to foot the bill.

    • Please supply links and graphs. Thank you! 😀

      • Jacob

        “Unemployment has never feel below 3%.” This is actually a good thing. Economists have found that the ideal level of unemployment is greater than 4% but less than 5.7%. This is because, I believe, when we have unemployment levels less than 4%, it could be a big indication that we are in a unsustainable expansion (the housing bubble was a good example of unsustainable expansion).

        Also, one must remember that there are 3 types of unemployment. They are as follows:

        1) Cyclical Unemployment
        2) Structural Unemployment
        3) Frictional Unemployment

        Cyclical is related to recessions/depressions–it is bad in the long term and short term.

        Structural Unemployment is related to when a technology or method cause unemployment–for example, when the plow was invented we needed less berry pickers (thus causing people to unemployed). This is definitely something good in the long run because those berry pickers find other employable opportunities in the economy thus making us produce more for our hour worked, etc.

        Frictional Unemployment is voluntary unemployment. Think guy who doesn’t accept the first available job because he believes he has better qualifications and wants a job that he can commit too and keep while being satisfied (nothing wrong with that!)

  • Anonymous

    Guess what countries are on the downslope of that first chart? Gabon, Nigeria, Benin, El Salvador. These countries are capitalist, yet they don't have the GDP base to pick up the slack for what government has to pay to protect its citizens.

    • Indeed. I would suspect that the failure to protect citizens is one of the primary causes of the lower than average GDP.

  • Anonymous

    look at that chart more carefully, the majority of countries at 20 %and 30% are actually outperformed by the clusters at 40%

  • Left out of your list of things that are appropriate for government to spend money on are such things as roads. do you really think that all the roads in the country should be privately owned, operated and paid for? If so, how would the Interstates, that Eisenhower authorized and are owned by all of us, have common standards and how would they be paid for?

    • Good question. I haven’t done enough research on the subject to comment intelligently so I’ll defer to you on that. 🙂

  • Shea Kauffman

    This doesn’t in any way address what the money is spent on. It’s well known that war spending has a negative long term impact on GDP, but infrastructure and food stamps have a positive effect on long term growth.

    • Excellent point. It would be great to have a different graph for each sector of government spending.

      I looked at your site. You’re kind of brilliant. I took my first programming class at Evergreen in 1999. Right before you went there! Neat 🙂

  • Shea Kauffman

    Awesome, thanks.

  • Andy

    “The reason GDP matters is that it is directly correlated to a reduction in human suffering. Economic growth is the primary differentiating factor between a life lived in the US and a life lived in North Korea. Economic growth is the primary differentiating factor between a life lived in the 11th century and the 21st century.”

    The fact that Denmark tops the list in satisfaction disproves these statements.

    And including Ireland as a poster child for high GDP growth–that’s hysterical.

    • I agree that other factors can produce outliers such as these. However, citing outliers resulting from other variables, does not disprove an overall trend. Please supply links and graphs showing an overall trend consistent with your views. Thank you! 😀

  • Davis

    Growth rate and Govt spending to GDP has a lot to do with more developed countries tending to have lower growth rates (Law of large numbers) but having large social safety nets and infrastructure. You need to do a better job of arguing government as the hindrance of growth rather than larger government existing as society develops.

    I don’t disagree with your statement, I just think you would need to isolate this in some way. As it stands I don’t think you can use it as an argument.

  • Steve Roth/

    1. More prosperous countries tend to grow more slowly and also have higher government spending/GDP. As others have mentioned, correlation/causation. Can be useful to correct for starting GDP to accomodate the notion of “convergence” — the notion that less-prosperous countries ought to be catching up.

    2. A lot of the graphs here are of pretty small and/or messy data sets, and of short or single-snapshot time periods. People love to cherry-pick periods, and economists don’t analyze or present multiple overlapping periods very effectively.

    3. Here’s a data presentation that overcomes #2 pretty well: It shows that over the last several decades europe has not caught up (it did hugely ’45-’75), but neither has it fallen behind. Europe taxes about 40% of GDP, the US 28% or less.

    4. Nijkamp and Poot’s 2003 meta-analysis looks at dozens of studies and suggests that government size in prosperous countries has an uncertain to nonexistent relationship to economic growth.;h=repec:dgr:uvatin:20020028 It certainly doesn’t give anything like the suggestion of certainty that you suggest here.

    • Hi, Steve Roth/ ! thanks for commenting! Does your 28% tax rate include state and local taxes? Does it include spending borrowed money? I would think you would want to look at spending as a percent of GDP rather than taxes.

      If you do, I believe you’ll find that Europe and the United States are about the same w/ respect to spending as a percentage of GDP. The US just spends more on militarism and Europe spends more on social services.

      With regard to cherry picking time spans, could you cherry pick a 10 year time span that shows a positive correlation between government spending and GDP growth?

      If you can, i would love to put it at the very beginning of this article. Thanks! 😀

  • I agree 100% with Acolin.Continuous growth is unsustainable and contrary to the known laws of physics.We live in a finite system of limited resources and growing population.With the exception of sunlight,it is a closed system and it’s entropy is increasing.’Growth’ is exactly the wrong priority,in the big picture.

  • DC

    The 2004 data is pre-crisis data. Ireland, the small government country which so skewed the numbers (Btw, 3 v 7? Why was the data set so uneven to begin with?), had a small government that was riding high on EU investment which allowed it to keep taxes low. The US and Ireland also crashed harder that the ‘big government’ countries (save perhaps Italy).

    One big reason why countries grow fast is that their markets aren’t saturated. India and China have grown incredibly fast for long periods of time in large part because they’re relatively undeveloped– and that’s with Chinese-style state capitalism in the MORE successful case. Developed countries have had relatively slow growth for a while.

    It may be a more tenable argument to suggest that government spending and economic well-being are closely linked. A prosperous economy allows investment in social programs and infrastructure which over time may result in a more robust economy (one better able to weather hard times– Scandinavia and Germany have certainly weathered the crisis better than the US in terms of unemployment).

    The other question is: Spending on what? The US spends far more on defense (which is spending on employment directly [soldiers, etc.] and indirectly [factory workers, etc.]) than any other country on earth. While that employment may be good for employment rates, it is questionable whether it constitutes a good investment. What kind of skills do the veterans have once they leave and enter the job market? How much do we spend on welfare for veterans that they wouldn’t otherwise receive? And on top of that, there is straight-up wasteful spending (e.g. the F-22) and war-related spending (also of questionable investment return).

    Basically, my points are: 1- a large data set would be more accurate, especially if you figure in developing nations and compare them with developed ones; 2- what does the spending go towards? If you can find a long-term relationship between high education, health, defense, etc. spending and positive or negative economic growth, you’d get a clearer picture (e.g., the Danish ‘flexicurity’ system is qualitatively different from, say, the American system in terms of government expenditures).; 3- demographics may also play a role (the US spends a lot on pensions because we have an increasingly top heavy demographic; younger countries [often developing ones] don’t do that if only because they don’t have to).

    • I agree that military spending is more likely to harm GDP growth.

      However, I’m just looking at general trends based on many countries over many years. You can always find outliers (like Ireland) when looking at statistical data.

      Like you said, there are many variables (such as how much goes to militarism) that can pull a data point away from the trendline.

      However, the trendline exists.

      Maybe it would be better to look at military spending as a percent of GDP vs GDP growth


      social spending as a percent of GDP vs GDP growth.

      If you can find any graphs of those, I would be happy to include them.

      THanks, DC! 😀

  • Nilanjan

    You try to argue that higher government spending is bad, but fail to look at the graph of measured satisfaction after doing so. It clearly shows that higher government spending countries have higher satisfaction levels. Your thesis fails, badly.

    • Please quote the section where I say “government spending is bad”. I cannot find that statement which claim is my thesis.

      I believe I stated: “The line drawn through the points indicates that a 10-percentage-point increase in government expenditures as a share of GDP leads to approximately a 1-percentage-point reduction in economic growth.”

  • lalala

    A well-functioning economy requires government to:

    enforce contracts
    protect a country from foreign invaders
    keep people from stealing from each other
    keep people from hurting each other
    Where’s the proof for this?

  • This is interesting data, but does not support the proffered interpretation. What it shows, is an ‘optimal level’ of spending, not a linearly diminishing return. Rather, below the ‘optimal level’ it is quite clear that more government spending is needed. But, this overall analysis is incomplete at best. It is also necessary to look at such factors as stratification and concentration of wealth and income. Inequity produces privation and rebellion. There is a case to be made that freedom and happiness diminish beyond a certain point; but also that they increase up to that point. It is also an obvious point that growth is going to be greater in economies that are underdeveloped, and less in those that are highly developed – because ecology imposes constraints, which are not considered in this analysis. So, in sum, the analysis is simplistic and naive, but makes a point – mostly that nations should strive for balanced and sustainable development, not maximum growth nor minimum government expenditures.

    • Hi, Richard H. Moore!

      THanks for commenting! 😀

      I agree that looking at correlations of GDP growth with those other factors would be optimal. Unfortunately, I haven’t found the time to do that yet.

      However, doing that would not disprove the fact that in general above 20% government spending to GDP ratio, GDP growth declines.

  • Dietrich Krueger

    Did anyone notice that all the European countries with relatively high government spending rank very high on their people’s satisfaction? Perhaps there is a connection left unanalysed??

  • Pingback: league of legends rp generator()


Send me an email and I'll get back to you, asap.


©2017 Mike P. Sinn

Log in with your credentials


Forgot your details?

Create Account