Big government doesn’t kill growth???

By Leith van Onselen

During the Federal Election campaign, Labor’s shadow treasurer, Chris Bowen, confirmed that the overall tax burden would hit 24.8% GDP by 2026-27 under Labor, up from 23.5% in 2019-20:

Mr Bowen told The Australian Financial Review that his number was lower than the 25.7 per cent of GDP that Treasurer Scott Morrison claimed Labor would deliver, but higher than the Coalition’s ceiling of 23.9 per cent.

Mr Bowen said the alternative would be spending cuts to essential services.

“Let me be clear: tax-to-GDP will be higher over the medium term under both the Coalition and Labor government. Either that, or the Coalition will continue to deliver more savage cuts to Medicare and education,” he said.

The admission was immediately seized upon by Treasurer Scott Morrison, who claimed that a higher tax burden would damage the Australian economy’s growth:

“Labor might want to think you can have a tax-to-GDP ratio approaching 26 per cent and that will have no impact on the Australian economy. They are kidding themselves…”

The Coalition’s 23.9% of GDP ceiling on tax is based on the National Commission of Audit’s recommendation that taxation revenue as a share of GDP should be capped at 24%.

The assumption that higher tax equals less economic growth is a popular one among conservatives, not just in Australia.

However, four American academics have published an important new book, entitled “How Big Should our Government Be?”, which examines in detail the vexed issue of government size and growth.

According to the Washington Post, which provides a good summary of the book, there is actually a positive correlation between the size of government and economic growth per capita:

Using data on 12 advanced economies from 1870, the authors of the book conclude with the following:

“In the century and a half since then, government expenditures as a share of GDP have risen sharply in these countries. Yet they didn’t experience a slowdown in their long-run economic growth rates. The fact that economic growth has been so stable over this lengthy period, despite huge increases in the size of government, suggests that government size probably has had little or no impact on growth.”

The authors also note that “A national instinct that small government is always better than large government is grounded not in facts but rather in ideology and politics,” and that the evidence “shows that more government can lead to greater security, enhanced opportunity and a fairer sharing of national wealth.”

In particularly, the authors call for more investment in infrastructure, education, as well as proper safety nets for the unemployed and those that get sick.

The Turnbull Government should take note as it considers taking an axe to Australia’s public services.

MY REBUTTAL:

Let me start by saying that I am not in favor of austerity as a response to a major economic slow-down. If anything that will exacerbate unemployment and prolong the contraction. Instead I advocate major infrastructure programs to stimulate the economy. But with two caveats: (1) investments must generate a market-related return on investment; and (2) there must be strong involvement from the private sector. Investment in assets that do not generate direct revenue leaves future taxpayers with a pile of debt and no income (or saleable assets) that can service (or repay) it. Involvement of the private sector should be structured to ensure that the benchmark of market-related returns is not superseded by projects selected to win the most votes. Also, the private sector should have skin in the game to restrict cost blowouts. They are not immune to cost blowouts but are not in the same league as big government.

I also believe that weak government will harm an economy. We need strong regulators, rule of law, police and military to ensure stability. Also spending on education and science to foster growth.

But the article by Jared Bernstein in the Washington Post typifies the kind of rubbish pedaled to voters around election time. And seems to have been swallowed hook-line-and-sinker by the author of the MB article.

Where do I start?
First, the fact that a book by four unnamed academics is cited as proof in itself should tell us how much credibility to attach to their claims.

Second, the author mentions that there is “a positive correlation between the size of government and economic growth per capita…”. A positive correlation is any correlation coefficient greater than zero. The highest correlation is a value of 1.0 which represents a perfect fit. No correlation coefficient is provided in either article and judging from the graph I would assume it is closer to zero than 1, meaning there is only a vague correlation. If you ignore the line drawn on graph, the data looks randomly scattered with no clear trend.

Also the author overlooks that he is only dealing with a sample of 12 countries, which again would give a low level of confidence in any result.

Further, in the WP article the author concedes that correlation is not equal to causation: “That positive slope in the figure on the left above could easily be a function of reverse causality: As economies grow, their citizens demand more from them.” This is omitted in the MB report.

Then the study of data for the 12 economies from 1870 up top the present is used to argue that growth in government expenditures does not hinder GDP growth. I would be surprised if the data didn’t show growth in government across all countries as it spans the era from horse-drawn carts up to the area of modern jets and space travel. From the country GP with a stethoscope to modern nuclear medicine and MRIs. From slate and chalk to super-computers and digital technology. Of course the demand for infrastructure has grown exponentially over that time. To argue otherwise would be stupid.

But that is not an argument in favor of a welfare state or increased government expenditure. In fact, most of those advances in technology were driven by private individuals and not by government.

Finally, I will use another graph from “How big should our government be,” Bakija et al in the same Washington Post article to argue the case for lean government (as opposed to small government circa 1870):

The graph shows that tax revenues as a percentage of GDP have steadily declined, since the late 1990s, for every country except France. Why has this occurred in even model welfare states like Sweden and, to a lesser extent, Canada? Simply because they reached “peak welfare” in the 1990s and realized that the only way to revive GDP growth was to reduce the role of government in the economy.

The only one who hasn’t accepted the evidence is France. Which may well be contributing to their poor economic performance.