Bringing Manufacturing Back to the U.S. Is Easier Said Than Done

Bringing Manufacturing Back to the U.S. Is Easier Said Than Done

by Willy C. Shih - April 15, 2020

The Covid-19 pandemic has raised a critical question: Why does the United States not have the capacity to manufacture many products for which there is a sudden urgent need - everything from critical care ventilators, N95 face masks, and personal protective equipment to everyday items like over-the-counter pain relievers? Of course, the United States is still a manufacturing powerhouse in many sectors, but it surprises many people that a huge number of everyday basic items have to be imported. The current pandemic-related shortages have fueled calls from political leaders of both parties for U.S. manufacturers to start producing critical supplies domestically. And long before the pandemic, President Trump was pushing U.S. companies to bring back production from overseas.

The issue is complex and defies easy solutions. The challenge lies in a combination of how modern supply networks are structured and the operational metrics applied to manufacturers. Taken together, the United States and other advanced industrial economies have evolved a highly efficient and productive product manufacturing-and-delivery system that provides them with a cornucopia of products at relatively low costs. But inherent in that system are dependencies and expectations that the pandemic has called into question.

Modern products require high degrees of specialization.

The days are long gone when a single vertically-integrated manufacturer like Ford or General Motors could design and manufacture all or most of the subassemblies and components it needs to make a finished product. Technology is just too complicated, and it is impossible to possess all the skills that are necessary in just one place. Consequently, manufacturers have turned to specialists and subcontractors who narrowly focus on just one area - and even those specialists have to rely on many others. And just as the world has come to rely on different regions for natural resources like iron ore or lithium metal, so too has it become dependent on regions where these specialists reside.

Even something as simple as an energy-efficient desk lamp has sophisticated components like LED lights that are made in high-tech factories. Devices like smartphones, medical equipment, and precision instruments contain components whose design and manufacture require a great deal of specialization. The design and manufacture of modern microcircuits involves sophisticated tools, and the people using them need considerable training and experience to operate them successfully.

The dependence on specialists is clear if we look at a typical notebook computer. Companies like Dell and HP rely on a handful of Taiwanese original design manufacturers to do the assembly work, but those assemblers, in turn, depend on multiple subsystem manufacturers.

For example, the display is made up of a number of components. At its heart is a thin-film transistor liquid crystal display (TFT-LCD) panel, which is mated with a backlight assembly and bezel. The TFT-LCD panels are made by a handful of Asian manufacturers in large, capital-intensive factories - the most recent of these cost more than $6 billion each to build and equip. These panel makers, in turn, are dependent on others who supply essential raw materials such as optically flat glass sheets, polarizing films, flexible circuit connectors, display driver chips, and a host of other inputs. The display driver chips are made in semiconductor factories (“fabs”) spread around the world.

Other key subsystems require similarly narrow skill sets. The memory chips are made predominantly by three global specialists in their multi-billion-dollar fabs, and the hard drives by two firms with factories in Thailand, Malaysia, and China. The microprocessor is generally made by either Intel or AMD. Intel produces chips in the United States and other locations, but sends them to Asia to be packaged. AMD has them made in Taiwan.

The long-term trend towards specialization in most fields is increasing because of the very different technological skills and capabilities demanded of firms working on the leading edge. Whether you are making computers, food ingredients, or personal care products, this division of labor helps firms incorporate new technologies and do so more economically than ever before. Specialists are also able to exploit scale economies both in production and design, making it harder for firms who might wish to become self-sufficient to perform those tasks economically.

The end result is that we have many suppliers scattered around the world upon whom manufacturers depend for critical components. Electronic product companies are heavily dependent on suppliers across Asia (primarily China, South Korea, and Japan). Those relying on industrial enzymes might have to turn to Denmark. Indian pharmaceutical makers rely on Chinese suppliers of active pharmaceutical ingredients. Many manufacturers have to rely on precision toolmakers in Germany, Switzerland, or northern Italy or robot makers in Germany or Japan.

A consequence of these complex interdependencies is a deep tiering of supply chains, with manufacturers dependent on their first-tier suppliers, which, in turn, are dependent on a second tier, which are themselves dependent on a third tier, and so on. Visibility into third, fourth, and more distant tiers is challenging, making wholesale replacement of anyone in the chain, let alone the entire chain, extremely difficult.

Thus, while it might seem appealing for President Trump to invoke the Defense Production Act and force automakers to pivot to manufacturing medical ventilators, it is very difficult for them to ramp up production if key components like pressure sensors or valves are made by an offshore specialist. Even something as simple as an N95 mask made in the United States by 3M uses (according to the label on the box) “globally sourced materials.”

Scaling up production requires unique skills.

A manufacturer not only has to source all of the components of a product, it also has to scale up production. This task is often taken for granted, but it is part of the really hard work of taking a product to market. The process includes setting up the supply chain for all of the raw materials, designing an assembly process with the appropriate tooling and fixtures, building or securing test equipment, establishing testing and quality procedures, and working through materials handling, staffing, and countless other details.

While there are many firms in the United States that know how to put products into production, their number is much lower than what it used to be. That is because the job of taking a product into manufacturing has increasingly turned into one of sourcing from offshore producers. One manufacturer that had offshored a lot of its production told me, “Operations management leadership has turned into procurement leadership.” Increasingly, the job has been to specify the product for an offshore original-design manufacturer or to transfer the work to a contract manufacturer. In an emergency, when a U.S. company suddenly needs to scale up, the skills are hard to find.

Manufacturers want to maximize utilization of plant and equipment.

One of the characteristics of modern manufacturing operations is a keen focus on operational efficiency. Factory managers might track a number of key performance indicators, all of which motivate them to size their operations with minimal surplus capacity.

Overall equipment effectiveness (OEE) is the percentage of time that a factory is truly productive. A score of 100% means that you are producing 100% good parts (no defects) as fast as possible and are never stopping production. In practice, most production lines schedule downtime for preventive maintenance or changeovers, so scores of 85% are considered good. But given their focus on OEE, managers are reluctant to install excess capacity. That means they size a factory to handle the expected demand, with some surge capacity but not a lot of excess capacity.

Capital efficiency - how much capital you have deployed in your business - is another thing that is important to shareholders and Wall Street analysts.  Nobody wants to pay for idle or underutilized capacity, and in sectors where the capital expenditures for plant and equipment are extraordinarily high (think semiconductors, flat panel displays, automotive assembly, materials processing), investors applaud the outsourcing or offshoring to someone who is willing to invest or to a geography where they can receive subsidies.

Another approach taken by many product companies is subcontracting production work. They might retain a base load of work internally and turn to contract manufacturers or outsourced manufacturing service providers for variable capacity or seasonal needs. This way they can keep their own plants fully loaded. The contracted suppliers, in turn, use demand pooling across multiple clients to smooth out their own workload and try to maximize capacity utilization, which is how they can achieve lower operating costs. But product companies are increasingly expecting their contractors to operate dedicated factories just for them, taking away the demand pooling benefit and forcing those contractors to keep capacity tight. This deprives the overall ecosystem of production flexibility when there is a sudden need.

The desire to avoid capital investments also leads to risk aversion to investing in new manufacturing technologies. I worked with a company that was supplying quantum dot backlighting technology for LCD flat-panel displays. Manufacturers were insistent that any new technologies had to fit into their existing capital-intensive workflows. I also heard a leader of a U.S. Department of Energy R&D group worry about how the extensive battery-manufacturing infrastructure in China had moved so effectively down the cost curve that it made it difficult for a potentially superior new technology from several MIT spinoffs to compete let alone raise the capital for a new production facility. The problem for the United States is exacerbated by countries like China that subsidize the construction and equipping of new production facilities.

The rise of lean production and its “inventory is evil” principle.

When Toyota first designed its Toyota Production System (TPS) in the decades following World War II, one influence was the company’s lack of resources to compete with well-capitalized large automakers in the United States. Its TPS lean production system was truly a revolution in manufacturing and was predicated on minimal inventory that was pulled quickly from suppliers located nearby. It has been replicated around the world in many different industrial sectors.

Efficient transport logistics have lulled major companies into building globally distributed, lean production systems. Pronouncements by industry leaders like Apple’s Tim Cook that inventory is “fundamentally evil” reflect the prevailing view that inventory along the supply chain not only risks obsolescence, it represents cash that is tied up that could be used for better purposes. So as companies have moved from a “Toyota City” model with suppliers clustered in a tight geographical area to supply chains that span the globe connected by dependable and predictable logistics links, firms have continued to squeeze inventory out of their supply chains whenever and wherever possible.

The end result is when we experience a supply shock or sudden disruption in raw materials, components, or whole product supply, there is little buffer inventory around to absorb that shock. When Indian drug manufacturers started running short of active pharmaceutical ingredients manufactured in China, the Indian government responded both by offering to fly materials in and restricting exports of finished products. It would have made more sense to carry six months of buffer inventory in a strategic stockpile.

The relentless drive to reduce costs.

Leaders of product companies go to their plant managers or their outsource manufacturing providers every year and ask for “greater productivity,” which is another way of saying, “I want it cheaper.” Many procurement managers are measured on how much productivity they can deliver every year, and their bonuses are tied to it.

These pressures lead to a race to the bottom in production costs in which product companies have minimal incentives to maintain production locations in high-cost locations or to worry about geographic diversity in production. And that behavior is encouraged by consumers and business end users. If an N95 mask sitting on a rack at Home Depot that is made in China looks equivalent to an adjacent higher-priced one made in the United States, consumers typically opt for the less-expensive one.

Many managers consider the global pandemic to be a “black swan” event, one which occurs so rarely that we can’t really price it or factor it into our thinking on production systems. Few are willing to pay the cost premiums that diversifying sources of supplies or carrying more safety inventory would entail. So when we have a supply shock or demand spike like we are currently experiencing, the surge capacity isn’t there.

But the current pandemic is not the only black swan event of the last 15 years. Arguably there have been several - including the 2008 financial crisis, the 2011 Tohoku East Japan earthquake and tsunami, the flooding in Thailand, and the U.S.–China trade war. The trade war got some firms to relocate some of their production out of China, but movement has been slow, and supplier risk is still largely undiversified.

Never Waste a Good Crisis

The pandemic has been and will continue to be a major shock to global supply chains and sourcing strategies. It is as if we suddenly lowered the level of the ocean and exposed all kinds of risks and obstructions that were previously hidden from view. Managers should use the unfolding disruptions to assess their supply strategies and initiate actions that will improve their resilience in the future. Some steps to consider include:

Plan to diversify sources for critical components and materials. This might include geographic diversification, either partnering with the same supplier or using second sources where economically feasible. As an example, Taiwan Semiconductor Manufacturing Company has spread its most important fabs across three science parks in Taiwan. Intel uses multiple fabs across the United States, Ireland, and Israel to produce its microprocessors. Many manufacturers are wary of the expense of duplicate tooling and the challenges in balancing production workloads across multiple sites, but they may wish to reconsider as more weak links are exposed.

Where diversification is not possible, reconsider what levels of safety stock or strategic inventory reserves are appropriate. Raw materials or intermediate goods that are earlier in the value chain are less costly to carry, and it may make sense to have larger reserves on hand. If it will take time to replace lost capacity, larger inventories of finished goods could make sense. Novo Nordisk, who makes a large percentage of the world’s supply of insulin at its facility in Kalundborg, Denmark, maintains a five-year inventory to supply critical needs in the event of disruption.

Examine logistics bottlenecks and plan alternatives. The current crisis is wreaking havoc with container shipping and air cargo and now is spreading to domestic trucking. To safeguard themselves against such disruptions, companies either need to have suppliers closer to their production locations and markets where products are consumed, or they need to understand where critical bottlenecks exist in their logistics systems and develop back-up plans.

Reconsider capacity-planning strategies for strategic commodities like medical supplies. This will likely have to be in collaboration with national governments, which may be willing to subsidize extra capacity by making purchases for a national stockpile. Alternatively, a government might subsidize surge capacity via something like the U.S. Department of Defense’s Trusted Foundry Program or Civil Reserve Air Fleet program.

The pandemic and trade wars together highlight the brittleness of our global supply chains and trading system. Managers should heed the lessons and build more resiliency into their operations.

Willy C. Shih is the Robert and Jane Cizik Professor of Management Practice in Business Administration at Harvard Business School.