The Wall Street Journal recently published an article entitled “Auto Makers Retreat from 50 Years of ‘Just in Time’ Manufacturing.” It noted that the Covid pandemic, the semiconductor shortage, and the burgeoning U.S. conflict with China were forcing American auto makers to rethink the lean manufacturing technique first developed by Toyota in the 1950s and subsequently imported into North America. Just in time maintains minimal inventories of parts and runs production lines on extremely tight schedules, which requires high levels of coordination with suppliers. Trump’s trade war, the growing U.S.-China rift, and the Covid pandemic have all delivered shocks to global supply chains, and it is no surprise that mangers are prioritizing resilience over efficiency and returning to stockpiling inventories of parts and materials.
All this has reminded me that the American auto industry never really understood the underlying purpose of just in time manufacturing, which is less about saving on inventory costs, and more about social trust and social capital.
Toyota’s lean inventories were part of a broader system of kaizen, or continuous improvement, whose intent was to deliberately make a manufacturing system more fragile so that mistakes and inefficiencies could be caught more readily and managers incentivized to fix them. If a part was defective, or was of low quality, assembly line workers could not just reach into the parts bin and pull out another one. Instead, each worker had a cord by their workstation that if pulled would bring the entire production line to a halt. This gave managers a powerful incentive to get things right from the outset—e.g., by working with suppliers to improve parts quality—rather than covering up their mistakes and having them caught only in the quality-control area at the end of the assembly line.
This system is highly dependent on social trust. Workers had to be trusted not to pull the cord for frivolous reasons, just as managers had to be accountable for acting on the information that workers gave them. They had to believe that they were part of a common enterprise. Just in time would not work in a factory with highly adversarial worker-management relations like Britain or Italy of the 1950s and 60s, when angry workers would be pulling the cord constantly just to screw with their bosses. The efficiencies of just in time were not only the results of smaller inventories; they had to do with the better flow of information within a high-trust work environment that led to continuous improvements in production processes.
Back in the late 1990s, I thought I had discovered a big difference in the application of just in time at Ford and General Motors. After some initial reluctance, the U.S. auto makers took a pilgrimage to Nagoya to learn the secrets of lean manufacturing, and adopted a version of it in their factories. At the same time, Ford had made an effort to cultivate a better relationship with the United Auto Workers union (UAW). It assigned a senior vice president to work with the union and refused for a time to source parts from Johnson Controls, which was involved in a bitter dispute with the UAW. GM, for its part, made no similar effort. Its entire North American operation was shut down twice, in 1996 and 1998, due to disputes with union locals that, given the just in time system, disrupted their tightly coordinated production schedules. This labor-management hostility was in keeping with the very un-Japanese way GM treated its suppliers in the 1990s under Ignacio Lopez, its supply-chain vice president. Lopez did the opposite of the trust-based “relational contracting” used by Japanese firms, relentlessly squeezing them for every penny and setting them against one another. Ford, I surmised, had understood the underlying role of trust in the Japanese system and taken it to heart in the way it was treating its workers.
Right around that time I was invited to give a talk to Ford’s senior management in Dearborn, Michigan, an audience that included the company's CEO at the time. I was excited to test out my theory of how they had understood the role of trust, unlike their rival GM. My talk however fell completely flat. I remember the excruciating elevator ride after it was over down to the basement garage with a number of executives who had been in the room, who rode with me in silence and failed to make eye contact.
I was apparently completely wrong about Ford’s understanding of lean manufacturing. Whatever their motives for improving relations with the UAW in that period, it was evidently unconnected with a systematic view of the importance of trust in the working of just in time.
It seems to me that there are currently three broad models of labor-management relations among developed countries today. The first is a high-trust model that still prevails in northern Europe, in which unions are given a strong role in setting wages, in return for greater management leeway in work rules and hiring and firing. Then there are two low-trust models. The first is the American one, in which a highly adversarial struggle has been largely lost by organized labor over the last 40 years (though not in the public sector). The second is the southern European one, in which militant unions have been able to veto labor market reforms on the part of governments, despite their falling numbers in the workforce. (This may be changing under French president Emmanuel Macron, who was able to win some significant concessions early in his presidency).
With Biden’s election, we now have an administration that is vocally pro-union. This is a welcome change, given the prolonged anti-union campaign that has been waged by corporations and their Republican allies since the beginning of the Reagan administration. But the outcome we should be aiming for is not just a redistribution of producer surpluses from owners of capital to workers in a zero-sum contest. The goal should be the building of a positive-sum relationship of greater trust. Germany achieved growth and export success in this century because its trade unions were persuaded to accept more flexible work rules for higher wages. They could do this because they did not see themselves locked in life-or-death struggle with their employers.
And as American corporations re-think just in time, they should consider its underlying purpose. Building inventories and increasing resilience is a necessary hedge today, but it should not become a facilitator of slack in production processes. Efficiency is dependent on the flow of information, and social trust is the lubricant that makes information flow more readily.
American Purpose newsletters
Sign up to get our essays and updates—you pick which ones—right in your inbox.Subscribe