This column is a lead commentary in the VoxEU Debate “The Economics of the Second World War: Eighty Years On“
During WWII, the US was the most likely country to have experienced an economic stimulus from increases in spending because so little of the war was fought on American soil. For decades people claimed that WWII was a fiscal stimulus that pulled America out of the Great Depression.
The official facts seem to fit the story. Defence spending rose from 1.4% of GDP in 1940 to over 37% in 1945 and the federal deficit rose from 3% of GDP in 1939 to 27.5% in 1943. Meanwhile, civilian unemployment rates fell from 9.5% in 1940 to below 2% from 1943 through 1945. Real GDP per person reached a wartime peak that was 67% higher than the 1940 level.1
Studies of the effect of war spending, however, tend to find relatively small multipliers for economic activity. Barro (1981) estimated that the multiplier for WWII spending for the nation was around 0.6. For comparison, a multiplier of 1 means that a dollar of WWII spending raised income by that dollar; a multiplier of 1.5 implies a rise of income of the war dollar plus 50 cents extra in spillover benefits. Barro’s multiplier of 0.6 suggests that the war spending crowded out about 40 cents of private economic activity for every federal government dollar spent.
Gordon and Krenn (2010) found higher multiplier estimates for military spending prior to US entry into the war. The federal government ramped up defence spending from 1.2% of GDP in 1938 to 1.7 in 1940, and to 5.1% in 1941. From the start of 1939 through June 1942, they estimated a multiplier of 1.8 because the economy was still suffering from high unemployment and unused capacity. When US manufacturing hit capacity in the latter half of 1941, the multiplier fell to 0.88. Barro’s multiplier of 0.6 suggests that the multiplier fell still more as war production replaced more and more of production for civilian goods.
More recent studies focus on the effects of WWII spending at the county and state level. Brunet (2017) found a state-level multiplier of only 0.25 during the war and suggested that the estimate for the national economy is only 0.3. Cullen and Fishback (2013) examined how military spending inflows into counties influenced the change in peace-time economic activity between 1939 (before the war) and the late 1940s and 1950s (after the war).
The primary effect was a population increase with small effects on economic activity per person. The finding was the same when Fishback and Jaworski (2016) examined even longer-term effects on the 1960s through 2010. Jaworski (2015) tested to see if the surge in war spending in the South spurred a long-run rise in southern manufacturing and found only a small impact. In all of these cases, the authors suggest that once the economy reached capacity, the war spending crowded out normal economic activity.
The crowding out of normal activity became more obvious after the war ended. Keynesians had predicted that the reduction in government production and deficits would lower income, which in turn would lower private consumption and investment. Instead, consumption and investment rose sharply in the absence of war spending (Higgs 1999). In addition to the crowding-out problems, there were substantial transition costs of converting factors to war-specific production at the start and then reconversion to peace-time production after the war ended.
The estimates of multipliers during the war are unlikely to be applicable to the US market economy during peacetime because the structure of the economy was drastically different. The US war economy was a quasi-command economy in which the government forced 10% of the workforce to join the military at compensation levels well below normal wages. The military had the first claim on all resources, as over 36% of estimated GDP was devoted to the production of war goods that would be destroyed, left behind, or mothballed. Production halted on automobiles, civilian housing, and most consumer durables. The military also had first claim on the materials for clothing, food, and other factors. This led to rationing of meat, gasoline, fuel oil, kerosene, nylon, silk, shoes, sugar, coffee, processed foods, cheese, and milk.
Make no mistake. The war-time production that made the US economy ‘the arsenal of democracy’ was a tremendous accomplishment. In a very short time span, the US economy produced 17 million rifles and pistols, over 80,000 tanks, 41 billion rounds of ammunition, 4 million artillery shells, 75,000 vessels, nearly 300,000 planes, and many more items and services for the war.2
But were Americans better off during the war than they had been before the US entered the war? Higgs (1992, 2006) provides ample evidence that they were not. For most Americans, the WWII experience might better be described as a sacrifice in which they gave up their normal consumption, worked longer and harder, and a significant share gave up lives and limbs to join their Allies in defeating Germany, Italy, and Japan.
Consumption per capita measured with official prices shows no change in private consumption between 1941 and 1944 but the estimate does not account for the declines in quality of goods, the extra costs of obtaining rationed goods, and the complete absence of other goods. Once the consumption figures are adjusted to develop better estimates of the true prices, the amount consumed per person was lower throughout the war than it was in 1940 when the economy was still climbing out of the Great Depression.
Fighting the war and building munitions required people. In the last year of the war, 18% of the combined civilian and military labour force were in the military and another 22% were producing munitions. The men and women serving in the armed forces were typically paid, in money and in-kind, the equivalent of about two-thirds of the earnings of an unskilled worker at home. Their activities were determined by orders of superiors 24 hours per day and many were sent in harm’s way. Over 400,000 died and another 670,000 were wounded. Those who survived slogged through horrific conditions and experienced events that scarred them for life. psychologically. For each death or severe casualty, there were likely several family members and close friends who mourned their loss of life or limb. Families were separated for long periods and their time together was overlaid by the dread of the next separation.
Workers on the home front worked more intensively. In manufacturing, weekly hours rose from 38 in 1940 to 45 in 1944. Night shifts became more common and workplace injury rates rose. War work disrupted many long-term plans by drawing teenagers out of school, women from their homes, and the elderly out of retirement. Pay rose but was restricted by wage ceilings, and many had to migrate to new cities to take advantage of the new opportunities.
To pay for the war, the federal government sharply increased tax rates. The average tax rate for top incomes rose to 90%. Further, the number of households paying income taxes rose six-fold. Even families in poverty had begun paying income taxes.3 Even though federal tax revenues rose to 20% of GDP in 1945, war borrowing led the national debt to more than double to 105% of GDP. In addition, earnings were ‘taxed’ by inflation that eroded purchasing power by 5% per year using official prices and 9% per year using alternative estimates.
Despite the sacrifices, many remember the war as prosperous relative to the Depression because everybody had a job and developed a sense of shared sacrifice to defeat the Axis. Some individuals did fare better. Blacks migrated north and west to better jobs. Industrial demand for women’s services rose during the war; despite a post-war fall, it remained higher than in 1941 (Shatnawi and Fishback 2018). With little to buy, people accumulated wealth through savings or bought existing housing, which fuelled the post-war boom delayed by the war.
In sum, WWII involved extensive sacrifices by Americans. Yet, these pale in comparison to the sacrifices in the rest of the world where 60 to 70 million died and refugees fled as factories, farms, and homes were decimated. The US would have been much better off economically had it never entered the war. The world economy would have been vastly better off had the war never started.
Barro, R (1981), “Output effects of government purchases”, Journal of Political Economy 89: 1086–121.
Brunet, G (2017), “Stimulus on the home front: The state-level effects of WWII spending”, working paper.
Carter, S B, S S Gartner, M R Haines, A L Olmstead, R Sutch and G Wright (2006), Historical statistics of the US, earliest times to the present: Millennial edition, New York: Cambridge University Press.
Cullen, J, and P Fishback (2013), “Second World War spending and local economic activity in US counties, 1939–58”, The Economic History Review 66(4): 975–92.
Field, A (2008), “The impact of the Second World War on US productivity growth”, Economic History Review 61.
Fishback, P V, and T Jaworski (2016), “World War II and US economic performance”, in J Eloranta, E Golson, A Markevich and N Wolf (eds.), Economic History of Warfare and State Formation, Springer Studies in Economic History.
Gordon, R J, and R Krenn (2010), “The end of the Great Depression 1939–41: Policy contributions and fiscal multipliers”, NBER Working Paper 16380.
Higgs, R (1992), “Wartime prosperity? A reassessment of the US economy in the 1940s”, Journal of Economic History 52: 41–60.
Higgs, R (1999), “From central planning to market, the American transition, 1945–1947”, Journal of Economic History 59: 600–23.
Higgs, R (2006), Depression, war, and Cold War: Studies in political economy, New York: Oxford University Press.
Jaworski, T (2017), “World War II and the industrialization of the American South”, Journal of Economic History 77(4): 1048–82.
Shatnawi, D, and P Fishback (2018), “The impact of World War II on the demand for female workers in manufacturing”, Journal of Economic History 78(2): 539–74.
 Government fiscal data, unemployment, and GDP throughout are from Carter et al. (2006: 2-83, 3-21, 3-24 to 3-26, 5-103 to 5-105, and 5-109).
 Field (2008) finds that productivity grew less during the war than before or after. There were positive spill-overs from war research, like radar and microwaves, but many military production processes were too costly to be transferred to market production.
 A family of four started paying taxes at $600, well below a $993 estimate of the poverty line (Carter et al. 2006: 2-663).