Long-term trends in manufacturing, farming, and protectionism

Peter Drucker was one the most influential management scholars of the 20^th century. In 2000 he wrote a special report for The Economist magazine to think about changes over the 20^th century and looking towards the 21^st.

The 20th century saw the rapid decline of the sector that had dominated society for 10,000 years: agriculture. In volume terms, farm production now is at least four or five times what it was before the first world war. But in 1913 farm products accounted for 70% of world trade, whereas now their share is at most 17%. In the early years of the 20th century, agriculture in most developed countries was the largest single contributor to GDP; now in rich countries its contribution has dwindled to the point of becoming marginal. And the farm population is down to a tiny proportion of the total.

Manufacturing has travelled a long way down the same road. Since the second world war, manufacturing output in the developed world has probably tripled in volume, but inflation-adjusted manufacturing prices have fallen steadily, whereas the cost of prime knowledge products—health care and education—has tripled, again adjusted for inflation. The relative purchasing power of manufactured goods against knowledge products is now only one-fifth or one-sixth of what it was 50 years ago. Manufacturing employment in America has fallen from 35% of the workforce in the 1950s to less than half that now, without causing much social disruption. But it may be too much to hope for an equally easy transition in countries such as Japan or Germany, where blue-collar manufacturing workers still make up 25-30% of the labour force.

… The decline of farming as a producer of wealth and of livelihoods has allowed farm protectionism to spread to a degree that would have been unthinkable before the second world war. In the same way, the decline of manufacturing will trigger an explosion of manufacturing protectionism—even as lip service continues to be paid to free trade. This protectionism may not necessarily take the form of traditional tariffs, but of subsidies, quotas and regulations of all kinds.

He continues in a section focused on manufacturing:

Between 1960 and 1999, both manufacturing’s share in America’s GDP and its share of total employment roughly halved, to around the 15% mark. Yet in the same 40 years manufacturing’s physical output doubled or tripled. In 1960, manufacturing was the centre of the American economy, and of the economies of all other developed countries. By 2000, as a contributor to GDP it was easily outranked by the financial sector .

The relative purchasing power of manufactured goods (what economists call the terms of trade) has fallen by three-quarters in the past 40 years. Whereas manufacturing prices, adjusted for inflation, are down by 40%, the prices of the two main knowledge products, health care and education, have risen about three times as fast as inflation. In 2000, therefore, it took five times as many units of manufactured goods to buy the main knowledge products as it had done 40 years earlier.

… The purchasing power of workers in manufacturing has also gone down, although by much less than that of their products. Their productivity has risen so sharply that most of their real income has been preserved. Forty years ago, labour costs in manufacturing typically accounted for around 30% of total manufacturing costs; now they are generally down to 12-15%. Even in the car industry, still the most labour-intensive of the engineering branches, labour costs in the most advanced plants are no higher than 20%…

Manufacturing is following exactly the same path that farming trod earlier. Beginning in 1920, and accelerating after the second world war, farm production shot up in all developed countries. Before the first world war, many Western European countries had to import farm products. Now there is only one net farm importer left: Japan. Every single European country now has large and increasingly unsaleable farm surpluses. In quantitative terms, farm production in most developed countries today is probably at least four times what it was in 1920 and three times what it was in 1950 (except in Japan). But whereas at the beginning of the 20th century farmers made up the largest single group in the working population in most developed countries, now they account for no more than 3% in any developed country. And whereas at the beginning of the 20th century agriculture was the largest single contributor to national income in most developed countries, in 2000 in America it contributed less than 2% to GDP…

At the time this was written back in 2000, Peter Drucker was forecasting that manufacturing output would double by 2020 (100% increase), but now that we have data we can see that from 2000-2020 US manufacturing output only increased 4% total. (The red line just emphasizes the year 2000 for comparison.)

Meanwhile, we can add in the percent of the labor force that is working in manufacturing. Manufacturing employment plummeted from 12% of jobs to just 7.5% of jobs during the same period! So American manufacturers are producing 4% more output with drastically fewer workers.

The stagnation of manufacturing output can best be explained by the explosion of globalization since 2000. As manufacturing technology has improved, it has increased output for centuries, and so while improving technology is the main reason for job losses, it cannot explain stagnant output. Globalization fits the bill as America’s trade deficit in manufactured goods has risen. As you can see in this graph, the manufacturing trade deficit in 2019 was almost triple its average value for the 1990s. As Americans manufacture less, we are also exporting more financial products and services, so this doesn’t mean that the overall trade deficit has almost tripled, but it does help explain why manufacturing output is stagnant.

Drucker continues:

The decline of manufacturing as a producer of… jobs changes the world’s economic, social and political landscape. It makes “economic miracles” increasingly difficult for developing countries to achieve. The economic miracles of the second half of the 20th century—Japan, South Korea, Taiwan, Hong Kong, Singapore—were based on exports to the world’s rich countries of manufactured goods that were produced with developed-country technology and productivity but with emerging-country labour costs….

The decline in manufacturing as a [share of national] wealth and jobs will inevitably bring about a new protectionism, once again echoing what happened earlier in agriculture. For every 1% by which agricultural prices and employment have fallen in the 20th century, agricultural subsidies and protection in every single developed country, including America, have gone up by at least 1%, often more. And the fewer farm voters there are, the more important the “farm vote” has become. As numbers have shrunk, farmers have become a unified special-interest group that carries disproportionate clout in all rich countries…

America’s trade unions have …become staunchly protectionist and declared enemies of “globalisation”. No matter that the real threat to manufacturing jobs is not competition from abroad, but the rapid decline of manufacturing as a creator of work: it is simply incomprehensible that manufacturing production can go up while manufacturing jobs go down, and not only to trade unionists but also to politicians, journalists, economists and the public at large. Most people continue to believe that when manufacturing jobs decline, the country’s manufacturing base is threatened and has to be protected. They have great difficulty in accepting that, for the first time in history, society and economy are no longer dominated by manual work, and a country can feed, house and clothe itself with only a small minority of its population engaged in such work.

Drucker goes on to use Mancur Olson’s theory of concentrated interest groups to predict that industrial unions will grow in political power as the number of union members shrinks just as what happened with the farm lobby, but manufacturing unions are face much greater political problems than farmers because unions (in the American tradition) have always been opposed by an even more concentrated interest group: management. In contrast, farm owners and farm labor is often one in the same because the farm owners provide their own labor, so there hasn’t been much divide between the interests of labor versus capital in farming. In corporate America, management is legally set up to be solely an agent for the interests of capital owners. Partly because the agents of capital have had even more powerful financial backing than unions, they have been generally been winning battles with unions for over a half century already and manufacturing union power has been on decline by every measure: membership, pay differential, absolute pay, legal rights, etc. So Drucker was wrong in fearing the growing power of unions. It has been management that has been the bigger (and growing) power in lobbying for protectionism.

This is because the protectionism of manufacturers has benefitted management & owners of protected firms far more than union members on a per-capita basis, so it is management that has the biggest interest in lobbying for protectionism, not the unions that Drucker fears. Drucker seems to think that owners want free-trade, but both unions and management always want the biggest possible profits for their firms and protectionism always boosts profits for a protected firm. The conflict between the unions and management is in how to divide the profits and America has had a stronger tradition of conflict between labor and management compared with most European nations so it has been hard for labor unions to work together with management on things like lobbying for protectionism which could generate profits that benefit both at the expense of the rest of the nation.