By Mark Harrison, Professor of Economics, University of Warwick. Originally published at VoxEU
Economic warfare was widely used in WWII. When one country blockaded another’s supply of essential goods or bombed the industries producing them, why did the adversary’s economy fail to collapse? This column, part of the Vox debate on the economics of WWII, reviews Mançur Olson’s insights, which arose from the elementary economic concept of substitution. He concluded that there are no essential goods; there are only essential uses, which can generally be supplied in many ways.
Mançur Olson (1932–1998) is best known for contributions to the political economy of collective action (Olson 1965) and of comparative economic development (Olson 1982, 2000). In earlier work, Olson also provided novel insights into the economic adaptation of countries to international conflict.
When one country imposed trade sanctions on another, blockaded its food supply, or bombed its war industries, why did the results so often disappoint or surprise? This question puzzled and frustrated civilian and military leaders on both sides in two world wars. Olson proposed that the answer lay in the elementary economic concept of substitution.
The possibility of economic warfare arose when a country’s economy was fully employed in the supply of war. The strategy of economic warfare was to weaken an adversary’s fighting power by attacking it, not directly, but through its supply chain. The tactics of economic warfare then aimed to block or destroy supplies of the commodities thought to be essential to the enemy’s war production or its war economy more generally. It was a tactical success if ships were sunk or factories were destroyed.
But strategic success was achieved only if the enemy’s fighting power was weakened as a result. Given tactical success, would strategic success follow? Olson (1962) argued that the link from tactics to strategy would generally be undermined by the adversary’s adaptation. The key to this response, he suggested, was substitution.
Allied economic analysis suggested that ball and roller bearings were ‘essential’ to the supply chain of German munitions (Bollard 2019). From August to October 1943, the US Army Air Forces systemically attacked and largely destroyed the small number of factories around Schweinfurt that provided around half of Germany’s ball-bearing capacity. While the cost in aircraft and crew was heavy, the observed effect on German war production was near zero (USSBS 1946: 4-5).
Olson noted several reasons. A high proportion of Germany’s existing supply of ball-bearings was used unnecessarily, where plain bearings would also do. Plain bearings were easily substituted when the supply of ball-bearings failed so that the much smaller range of truly essential uses could still be met. In addition, to assure the essential uses, capital and labour were quickly diverted from other employments to rebuild the essential capacity in dispersed, less vulnerable locations. Thus, the German economy under attack was re-optimised for war by sliding along its production frontier, although at a cost to other less-important objectives.
This led Olson to be critical of model-based approaches to target selection (such as Wassily Leontief’s input-output framework) that assumed fixed coefficients in production and consumption. Such models implied that to deprive an economy of a single ‘essential’ commodity, whether ball-bearings, oil, or molybdenum, would be a crippling blow. But this followed entirely from ruling out substitution, which turned out to be crucial to the outcome.
In The Economics of the Wartime Shortage, Olson (1963) generalised his idea. He asked how Great Britain, of all nations most dependent on international trade, survived three major conflicts—the Napoleonic War and two World Wars—without famine. Olson noted that food was widely thought of as an ‘essential’ good and that, in all countries, food security loomed large in thinking about war preparations. This was the thinking of German leaders in two world wars when they applied submarine warfare to the blockade of the British Isles, aiming to cut the UK economy off from its main sources of food.
Olson rejected the idea that, in an integrated market economy, any one commodity, even food, was more essential than any other. At the margin, where choices must be made, the strategic value of a dollar’s worth of food would always be about the same as a dollar’s worth of anything else. In a rich society, food would have many uses, some essential and some inessential or luxurious. “It is not the type of good”, Olson wrote (1963: 9), “but the type of use that distinguishes a necessity from a luxury” (my emphasis).
Before WWII, Britain imported more than three-quarters of wheat and flour, oils and fats, butter, cheese, and sugar (Hammond 1951: 394). The Battle of the Atlantic was hard fought and very costly to both sides. By 1942, as Table 1 shows, food imports were running at just half the rate of the first nine months (October 1939 to June 1940). The loss of imports was only partly mitigated by a substantial increase in home production. Yet, after a dip at the end of 1939, British food stocks never fell below the pre-war level.
Table 1 British food supplies and consumption in WWII
Sources: Food imports and stocks are from Hancock and Gowing (1949: 206-207, 357-358); home production and energy consumed from Hammond (1951: 387, 393).
Notes: The figure for food imports under 1939 covers October 1939 to June 1940, and that for 1940 covers July to December 1940. The figures for pre-war home production are averaged over 1936-1938. The figure for pre-war food stocks is from the end of August 1939.
Most importantly, Table 1 shows the calories consumed per person remained essentially constant throughout the war, while their distribution was probably somewhat equalised by rationing. Rationing covered ‘luxury’ foods, but bread and potatoes were the most important sources of calories. These were never rationed, which also speaks to the adequacy of the food supply (Hammond 1951: 388). As for health, in 1942, deaths among children and adult civilians fell below the rates of 1939 and continued along the pre-war downward trend (Titmuss 1950: 521, 524).
Thus, Britain survived blockade despite initially relying on foreign sources for nearly two-thirds of calories for human consumption. Other countries that entered the war more nearly or entirely self-sufficient struggled and sometimes failed to feed their populations. They failed because they were poorer and so had fewer inessential uses of food at the outset or because their economies were insufficiently integrated so that efficient substitutions did not take place—or both.
The implications of Olson’s thinking were at the time, and remain today, contrary to the thinking of nearly all government leaders and advisers in every country, including Britain. For two centuries, the threat of war has prompted calls for a larger agriculture (or manufacturing industry), more food and oil security, and larger stocks of ‘essential’ goods. Any suggestion that the pursuit of self-sufficiency in such commodities is unnecessary, or even harmful, appears to lie well beyond the bounds of ‘acceptable’ discourse. Yet historical investigation shows that such efforts were often, if not always, misdirected.
It is tempting to swing the other way and conclude that economic warfare was always pointless or had no effect on the outcome of the war. Olson (1962: 313) took pains to reject this conclusion. He emphasised that supply-chain disruption was ineffective mainly when the economy was wealthy (so any commodity had many inessential uses) and when the commodity concerned was only partly interrupted (so enough remained for essential uses). He maintained that substitution had its limits.
As an example of when those limits were breached, he gave the German synthetic oil industry in 1944–45. Germany had no natural oil reserves and the pre-war creation of a synthetic oil industry was itself a substitute for a commodity in short supply. Access to Romania’s oilfields was lost in August 1944, making Germany entirely dependent on domestic sources. Repeated bombing of the oil plants in the summer of 1944 permanently reduced supply below consumption. By the time of the Ardennes offensive of December 1944, German plans relied on capturing Allied fuel stocks for their success (USSBS 1946: 8-9).
Four extensions are suggested. One is to the uses of economic assistance from one ally to another in wartime. During the decisive years of the war, the US economy, being twice the size of the combined economies of the UK and USSR, showered $50 billion of military-economic aid on Britain and the Soviet Union through the Lend-Lease programme. The framing purpose of Lend-Lease was “further to promote the defense of the US”—and nothing else. But that is not necessarily how the aid was used.
Inter-Ally aid turned out to be the converse of economic warfare. Just as the architects of the Combined Bomber Offensive did not predict and could not control the substitutions that the Germany economy made to adapt to destruction from the air, so too the US Lend-Lease administration did not predict and could not control the Soviet economy’s adaptation to the inflow of Allied munitions and war goods.
These resources were provided strictly to support Soviet fighting power. Because the external resources were at least partial substitutes for home resources; however, the Soviet authorities were able to respond by diverting those home resources to consumption and investment (Harrison 1996: 139-146). The re-optimisation described here was also an element in Olson’s later work (Olson and Zeckhauser 1966) on the free-riding problem in NATO.
Another extension is to the sources of national feeling in wartime. The effect of economic warfare on the enemy’s fighting power is indirect; it works via the economy. It follows that economic warfare always does ‘collateral’ damage to people who are civilians, whether or not they are part of the enemy’s supply chain. The result is often to stiffen the enemy’s resistance. The collateral damage inflicted on British cities by German bombers stiffened British resistance; the same done to German cities stiffened German resistance. The collateral damage of Germany’s submarine war on Atlantic shipping in WWI brought America into the war against Germany.
More generally, war is polarising and economic warfare extends that polarisation to the civilian population. This then facilitates what Olson saw as the enemy’s adaptation to economic warfare: economic warfare makes angry civilians more willing to tighten belts and make do with substitutes that would be unacceptable in peacetime. This does not make economic sanctions pointless, but it is a predictable consequence that should be reckoned with beforehand.
A third extension addresses the question: can economic sanctions be a substitute for battle? International relations since 1945 have provided many cases of economic sanctions aimed at forcing states to change their behaviour without bloodshed, most of them apparently unsuccessful (Jones 2015). Examples range from the Warsaw Pact countries in the Cold War to China, Cuba, North Korea, Southern Rhodesia, South Africa, Myanmar, Iraq, Iran, and Russia. In a few cases, sanctions or the threat of them have had completely unexpected side-effects: in 1941, US oil sanctions precipitated Japan’s attack on Pearl Harbor, while the fear of blockade was a factor in Hitler’s plan to seize the farmlands and oilfields of the Soviet Union. These examples suggest that economic sanctions may not ultimately save soldiers’ lives. They may achieve their goals only when backed up by the credible threat or use of superior fighting power.
Finally, Olson’s idea may be useful in illustrating the importance of economic analysis. When you teach the principles of consumer choice, consider whether your students may find the life-or-death consequences of substitution in a besieged economy to be a more impressive motivation than doughnuts versus pizza.
See original post for references
Olson is ignoring a number of issues.
In Germany, the ball bearings were far from the most important thing, true. But Germany was running into way more significant problems, amongs them notably transportation. The decimation of the German (more widely, West-European) rail network by Allied raids was massive, and Germans had rail problems already before. The destruction (especially from 44 onwards, especially using long range fighters and fighter-bombers that could target trains with high precision) of the German logistics was crippling. And there was no substitute. Similarly, there was little substitute for molybdenum and wolfram, needed for quality armour (or, for that matter, Swedish iron ore, which was way better than German one). There was no realistic substitue for the rail tranport in Germany.
In the UK – yes, the UK could provide the food. But at the expense of lots of other things (i.e if you’re growing food, you can’t do other things. Which si why the food part of L&L was so important for the USSR).
Perharps even more importantly, sinking the merchant marine was sinking the navy you’d need to trade. If Germany was able to sink the ships faster than you replaced them. From July to October 1940, U-boats sunk 282 ships. That is more than two ships a day on average. Compare to the Liberty ships productions later on, the average was three ships every two days. So still losing proposition to two ships a day sunk. No ships, not trade.
Crucually, he also commits the mistake a lot of economists do with the “oh, it will be just substitution” assumption. Any substitution takes time. In war, time matters. By the time you substituted, you might have well lost. (this is relevant to moder economies too, as the time and effort it takes to substitute at similar quality and quantity is often just waved away in the “assume can opener” way).
The temptation to refight the war is great!
Then there is Albert Speer’s claim that Germany wasn’t on a war footing (fully mobilized) until 1944.
In late ’44 the bombing of German railheads was found by Enigma intercepts to have brought rail transport to a standstill. But the advocates of attacking oil supplies won the priority debate—(based on my recollection of a piece in the NYRB. Can anybody corroborate?)
Allied planners in WW2 thought that the Germans were so super-efficient, that their economy would be ‘as tight as a drum’. A precursor of the just-in-time economy if you will. That is why the attacks against the ball-bearing factories. As it turned out, after those factories were hit the German planners started to work the phones and discovered that there was so much slack in the system, that they they had over a months worth of ball-bearings ready to go which would last them while they rebuilt the factories. Slackers!
The rail problem that vlade mentioned was made worse by the fact that the trains to the concentration camps had priority which led to German troops spinning their wheels while waiting for train transport as train-loads of civilians went sailing by. This was made worse by the German practice swapping units between the eastern and western front for whatever reason.
Olson may say that products can be simply substituted but it does not mean that it was a successful substitution. As an example, by the end of WW1 German troops were forced to use bandages based on paper that had been substituted for cloth bandages as there was no choice. Tough luck if you were wounded and needed a good bandage. Economic warfare is brutal and we saw this in modern times Some 500,000 children in Iraq died due to it but ‘the price was worth it’. Thousands have died recently in Venezuela too so how are you going to substitute for these deaths?
Frankly I suspect that Olson is not getting a true reflection of the reality of the situation with blockades. Jerome K. Jerome, who spent time in Germany before WW1, in his autobiography mentioned people like little old ladies he had know that had starved to death in the British blockade. Would the UK print stories of people starving due to the German blockade? You might have to read a lot of autobiographies of people who lived through those times to get a true picture of what was going on.
In any case, I believe that somebody ran the numbers on economic blockades and found that over the past century, that they do not work no matter how many civilians that they ended up killing. Modern day Yemen is proof of this.
On economic sanctions see the roughly 300 case studies that form the foundation for: Hufbauer, Gary Clyde, Jeffrey J. Schott, and Kimberly Ann Elliott. “Economic Sanctions Reconsidered 3rd edition (hardcover+ CD).” Peterson Institute Press: All Books (2008).
I have the impression that the single most devastating consequence of aerial disruption of the German rail network was the interruption of coal deliveries to power plants. Not sure where I read that, perhaps in Adam Tooze’s “Wages of Destruction”, but it may have been another author.
If that’s right, there was one or more single-points-of-failure within the Nazi economy, but it took most of the war to locate it.
Ball bearings are a tough case.
There are a number of metrics that show that the Russian’s production quality of their tanks fell off during their crises period and improved somewhat toward the end of the war. There is also the same, but I have not seen it in as clear cut a form, shown for the German’s. Since they didn’t survive their crisis, quality dropped and never recovered.
Ball bearings would be part of the “quality” issue, but very hard to quantify. The reliability of a lot of the German tanks was never particularly good because of rushed design. As the tanks were loaded up with bigger guns and armor, but using the same engines, overloading issues were added to the rushed design of their later designs. Somewhere in that mess, you are trying to get by with using less ball bearings.
If I am teaching a course and trying to find a good example of substitution that can probably be quantified. I might look at the (coal-based) synthetic oil issue. This is a case where you can probably get descent numbers. And you can look at the reduction of oil imports versus the added cost of the synthetic fuel.
You could also look at what part of the rest of their economy were the Germans willing to sacrifice to keep the program running.
If there are “no essential goods, only essential uses” we might want to insure that our uses are made sustainable. And for basic survival because modern life is not survivable let alone modern warfare. Foot soldiers and tanks are no longer an essential use of force. Neither are planes. Nor trains. Don’t tell all the would-be belligerents, but we’ve all got hypersonic nuclear missiles that can travel half way around the world. We’ve got a redundancy of satellites. And modernized grids. Bio warfare. Weather manipulation. We’ve got mass destruction down pat. The word “blockade” is a punchline. We’ve gone MAD. But one small problem, we’ve got no where to run. So clinging to the patriotic hope of a long drawn out fight to be victorious is as silly as it gets. There won’t be any way for “substitution” in a time of war. It’s nauseating to think about World Wars. But it is encouraging to think we can substitute neoliberalism for an economy of collective action going into the future. I’d say first on a national scale. Only substituting when disaster prevents a good harvest, etc.
This analysis makes the same mistake that writers make about US war on terror claiming it is failing, not winning, chaotic, etc.
The relatively recent goal, 20+ years or so, with sanctions as well as military exploits is demand destruction. I read somewhere that the West has some 8000 sanctions in place around the world and the military has bombed how many Countries “back to the stone age? How much would all these Countries be consuming if none of this was taking place?
Substitution was only a big factor back in the good old days of plenty. Now we need to substitute less for more.
With all due respect, humans are complex machines requiring more than just raw fuel (calories).
It’s the quality of the nutrition, ie protein, vitamin complexes, that determine critical health factors.
A diet poor in those will have serious deletrious effects, including inhibiting cognitive development.
That an equal calory regimen was maintained during wartime rationing is commendable, but it says
nothing about the actual levels of penury.
I seem to recall from a Military Science lecture years ago that military science found a weak part in Japanese tail assemblies that resulted in great vulnerability.
What about the U.S. cut off of scrap iron and the blockading of 100% of essential imported oil to Japan pre-Pearl Harbor?
As to the “surprise attack”, Japanese Naval officers designed an attack on Pearl Harbor as part of their studies. Would the U.S. deliberately allow Peal Harbor to be bombed? Looking at our history of interventions, coups, assassinations and the lies told to the American people by Washington,
“As early as 1927, war games at the Japanese Navy War College included an examination of a carrier raid against Pearl Harbor. The following year, a certain Captain Yamamoto lectured on the same topic. By the time the United States moved the Pacific Fleet from the West Coast to Pearl Harbor in May 1940, Yamamoto was already exploring how to execute such a bold operation. According to the chief of staff of the Combined Fleet, Vice Admiral Fukudome Shigeru, Yamamoto first discussed an attack on Pearl Harbor in March or April 1940.”
There is that little issue of Japan using American oil and iron to conquer in a quasi genocidal manor all of China.
Although it looks like someone in the State Department was slanting the words of the communications between the two governments, no one in the United States’ government realized that Japan thought it only real choices was to be allowed to keep invading China or else to go to war with the United States, as well as the British, French, and Dutch at the same time. The goal was to stop the Japanese invasion and maybe retreat back to Manchuria.
Back in the mid-80s when I was in grad school doing international relations Olson came to do a couple days of guest lectures. He had an article applying his collective action stuff to IR and was the focus of his talk. In it, in the formalization part of the article, he very clearly had a term that he interchangeably treated as either constant or variable as it suited his narrative. Trouble was, the math didn’t work out if the term was variable, and the the entire edifice of his theory demanded it be constant. And then he reverted to treating it as variable for the rest of the narrative part of the article all while claiming it fit a theory based on it being constant. Quite aggravating. I don’t understand how it got by reviewers. Sorry I can’t be more specific, I’ve tossed out nearly all my photocopies from those years. The clumsiness of this supposed great scholar has stuck with me all these years.
For an enemy not to be an immediate and total walk-over, it has to have a lot of spare capacity or what could be termed “resilience”.
In a truly contested offensive war, the initial belligerent is motivated by the ideology of a very small power-elite. Until that elite has been reduced or eradicated the “civilians” will suffer long and hard. For measures such as blockade to show results, the reactive adaptions inherent in the resilience that allowed for offensive war in the first place, have to be wrung out. Usually morale falters before total collapse happens unless the power-elite is a death-cult.
As to the current situation: mutually assured destruction is the shortcut to quick eradication of power-elites with ideology of any type, hence proxy wars and live military exercises such as the US in Iraq are the order of the day – it being less dangerous to have the most transgressive elements of the military playing away rather than at home.