Since the late 1970s it has largely been the consensus that “big government” is detrimental to growth. This manifested after the financial crisis when countries, including previously fiscally-comfortable countries like Germany and the UK, adopted austerity programs, ostensibly to spur growth by cutting government expenditure.
But our research shows the story is not so simple. We found that studies tend to reflect selection bias. Findings that indicate a negative association between government size and growth are more likely to be published than those that show either a positive or no association.
Our research also found that the affect of the size of government is different between developed and developing countries and that there is a lot we don’t know about the optimal size of government, and whether some parts of government should be smaller than others.
The existing research is inconclusive
The existing research on the effect of the size of government on economic growth is actually contradictory, with some researchers asserting that a bigger government enhances growth, and others arguing that it hurts growth.
The arguments for a positive impact of a big government rely on examples like the potential of infrastructure development to create jobs, or intervening when there is a market failure (e.g. taking over banks during the GFC). Some of the negative affects of a large government are thought to be felt through the excess burden of distortionary taxes, and government inefficiency.
But the research is ambiguous and inconclusive. A survey of the academic literature suggests the conflicting results could be a result of the decisions researchers make. About what measurement of government size is used, for instance, or the type of countries studied – developed or less-developed, rich or poor.
Our research considered these distinctions, and so we sought to account for these variations when examining the relationship between government size and growth.
We did this through meta-analysis - statistically analyzing 799 estimates reported in 87 existing studies, looking at the relationship between government size and economic growth. We looked at the different measures of government size (e.g. total government expenditures and government consumption) and different levels of development (developed and less developed countries).
We found only partial support for the idea that the size of government has an effect on economic growth. Specifically, our research suggests that the effect of government size on economic growth is negative in developed countries but insignificant in less developed countries (LDCs).
Put differently, while we find evidence of a negative effect of government size on economic growth in developed countries, we find no effect in the case of LDCs. This is the case irrespective of whether government size is measured as the share of total expenditure or consumption expenditure in GDP. It also suggests that big government is usually bad for growth in developed countries but not in LDCs.
What this means
There are a couple of things to take away from our research.
For starters, as has been hypothesised previously, a small government can enhance economic growth by providing the minimum for investment and growth - the rule of law and protection of property rights etc. But when an economy becomes richer, the size of the government tends to grow beyond its efficient level, so a further rise in government size would reduce economic growth.
This is explained by Wagner’s Law, which suggests that when a country becomes industrialized and richer, it will be accompanied by an increased share of public expenditure. But while there are certain forms of government spending which are necessary to sustain a functioning economy, spending beyond a specific level can bring more costs than benefits.
But the existing literature does not explain much about the optimum government size. Theoretically, there is a point beyond which increases in government size lead to a decline in economic performance. But empirical work is limited and inconclusive in this area, and thus it is not clear what this point is.
The distinction between developed and LDCs is also very important. Caution needs to be taken when generalizing the effects of government size on growth.
There’s a lot that we don’t know in this space. Further research is needed on the relationship between the size of particular parts of the government, and economic growth. Such studies are more likely to produce policy-relevant findings compared to studies that focus on total measures of government size.
This would help policymakers determine how big governments should be and which components of government to cut in the context of tight government budget constraints and excessive government expenditures.