World defense spending amounts to $1.63 trillion, or 2.6% of world GDP. Spread over the same time period as WW2 this is equivalent to 95% of the absolute cost of WW2 (using the GDP deflator to standardize value). In terms of share of GDP global military spending is about 1/6 of the level of military spending for WW2.
Every country has an army (except Costa Rica) and some countries devote an extraordinary large proportion of their resources to maintaining one. WW2 mobilized 4% of the world’s population and participants devoted 32% of their GDP to the war. According to the Stockholm International Peace Research Institute North Korea spends about 25% of its resources on its armed forces and has 5 % of its population on active duty. The US alone accounts for 42.8% of global defense spending and with China account for 50.1% of all military spending. Other countries spend quite little. About 20% of the world’s nations spend 1% or less of their GDP on defense and about 9% spend 3% or more of their GDP on defense. So a significant fraction of the world values defense at least 3 times as much as another significant fraction. Which of the two approaches is the wiser – hawks or doves?
One way to approach this is to look at the correlation between the mean IQ of each country’s citizens and defense spending as a percentage of GDP. It turns out to be zero. Another way is to look at opinion within countries. Ideally one would like to see how opinion varies with IQ within low defense spending countries and high spending countries. If there is a trend for the brighter citizens in both groups toward high defense spending intelligent option would be for being well prepared militarily. If the trend in both groups is for the smart to support less defense spending then pacifism may be the intelligent policy. If there are different patterns in each group then there is either an optimum degree of defense spending or it may be smart for the well armed to continue their arms race and for the rest to drop their arms altogether. Unfortunately I only have data for opinions of defense spending by IQ in one country – the USA. I can however use this data to answer part of the question.
The USA is one of the high defense spending countries – at 4.7% currently it spends roughly 4 times the global median on defense as a percentage of GDP.
Lets have a look at the Smart Vote says. Recall that the Smart Vote is based on the ratio of the proportion of people with IQs greater than 120 and the proportion of people with IQs less than 80 who support a particular choice. The ratio is multiplied by 100 to get rid of the decimals. The Smart Vote is the choice with the highest ratio at any point. All the opinion and IQ data comes from the General Social Survey.
Firstly the Smart Vote is decidedly for being interested in military policy.
Secondly in 1984 the Smart Vote expected an escalation of the arms race to be moderately likely, a reduction in nuclear arms to be moderately unlikely and the elimination of atomic weapons to be very unlikely before 1994.
Thirdly in 1984 the Smart Vote expected an all out nuclear war to be highly unlikely, a conventional war to be moderately likely and repeated guerilla wars to be highly likely before 1994.
Fourthly the Smart Vote didn’t expect the US to be involved in a conventional war between 1976 and 1992.
Fifthly in 2000 the Smart Vote considers terrorist threats from US citizens to be the same or larger than 10 years ago. All these expectations proved accurate.
Nonetheless in 1984 the Smart Vote considered military service in war or peacetime to be moderately important but not at all an obligation and before 1973 was for taking part in anti-war protests.
In spite of being against the use of conscription, and pro the exemption of conscientious objectors, the Smart Vote concludes that an all volunteer military hasn’t worked out well - somewhere between “Worse” to “Only Fairly Well”. The Smart Vote on Confidence in the Military over time confirms this. One can see in the graph below that the Smart Vote has been for “Hardly Any” confidence in the military up until the early 90s and since then somewhere between “Only Some” and “Hardly Any” confidence. The Stupid Vote (the lowest ratio of smart to dull opinion) has consistently been for “Great Deal” of confidence in the military. On the other hand the Smart and Stupid Votes have been converging a little over the last generation. In a related question the Smart Vote was for not being proud of America’s military record.
What about the question we started with – military spending? The graph below shows the Smart Vote on more, unchanged or less defense spending over the years and compares this to actual defense spending trends. The defense spending figures come from US budget history. I converted defense spending to an index by dividing the percentage of GDP spent of defense by 5% and multiplying by 100 to get rid of the decimals.
You can see that the Smart Vote has been consistently for less defense spending and that until the mid 1990s the Stupid Vote has been for increased defense spending. How much less spending? A question on how much less or more military spending people prefer was asked in the 1990 and 1996 and revealed that the Smart Vote is for much less military spending rather than just moderately less.
The Smart Vote ratios might reflect interests that just happen to coincide with IQ, in which case the above wouldn’t be intelligent opinion so much as a bias. I checked that possibility with a multiple regression. More versus Less Military spending was the dependent variable and IQ, political ideology, age, sex, race, date, income and education the independent variables. The table below shows the strongly significant independent relationships and their direction.
As expected ideology has a very strong effect on preferred military spending. Conservatives want to spend more and liberals less. Also preferring to spend more on the military, again as expected, are older people, whites and the wealthy. An unexpected finding was that women do too. There is a trend toward favoring ever less military spending over the years. Time, race and gender effects are very weak.
The important thing for the Smart Vote is that the relationship between higher IQ and favoring less military spending is still plain even after controlling for all these interest variables (that do in fact influence opinion). So the Smart Vote on this issue is less likely to be an artifact of bias and more likely to be a function of real insight and wisdom. Note that higher education leans toward less military spending independently of IQ and income. I interpret that to mean that not only does higher individual intelligence predispose one to want to reduce investment in the military but so does exposure to the intelligence and learning of others.
Intelligent opinion within the US has been accurate in its expectations of global military developments but nevertheless had little confidence or pride in the military, and believes that America should reduce its investment in military capacity substantially.
Given these consistently dove like preferences one could reasonably ask whether the Smart Vote is for outright abolition of the military, like Costa Rica. One way to guess at this is to regress the variations in Smart and Dumb Votes onto the variations of actual military spending over time. I did this after smoothing the saw tooth pattern in the Smart and Dumb Vote graphs above.
The correlations were -0.695 for the Dumb Vote ratio and 0.724 for the Smart Vote ratio.
The regression equations were
-0.848*Spending Index +143.85 = Dumb Vote ratio, and
0.824*Spending Index + 99.4 = Smart Vote ratio.
The Dumb Vote ratio is expected to exceed the Smart Vote ratio i.e. spending more will become as intelligent an option as spending less, when the spending index drops below 26.6. This is equivalent to spending 1.33% of GDP on defense. So the answer is no, zero defense spending isn’t the intelligent option for the US – there is an optimum at 1.33%. This amount of spending will still exceed the defense spending of China, Russia, North Korea, Iran, Iraq and Afghanistan put together so the US will hardly be seriously threatened even if it did reduce its spending to less than 1/3 of what it is now i.e. from 4.7% to 1.33% of GDP - $500 billion less.
An important question is why the Smart Vote is for less military spending in the US. The attitude of the US to international organizations has some relevance. A preference for less military spending tends to go along with a preference for the US staying in the UN and playing an active role in international affairs; a belief that the UN has too little power and that international organizations do not take too much power from the American government; and that the US should not follow its own interests if this were to lead to conflicts with other countries. Preferring less military spending also goes with believing that the UN should intervene if human rights are seriously violated. The Smart Vote is decidedly for all those views. That points toward the desirability of the US shifting away from isolationism, or American exceptionalism, which require bigger military defenses, and toward greater diplomatic and cooperative ventures, which would require a level of arms more in line with international practice.
Intelligent opinion in America thinks the US should try to be less offensive and more cooperative in international relations.