By Douglas Clement
Tom Holmes unearths the hidden gems in his economic research
• What factors influence the diffusion of Wal-Mart stores?
• Why is there such a high geographic correlation between nursing-home unionization and coal-mines?
• How important were railroads to the growth of the nation?
Through innovative synthesis of economic theory and data, Tom Holmes digs into such questions and unearths truths about how economies arrange themselves: a tradition of economics known as industrial organization.
"I am fundamentally interested in how production is organized," says Holmes, Curtis L. Carlson Professor of Economics. "I'm interested in seeing how big the productive units are and where they are. Do they put everybody in one big plant or lots of little plants scattered around the country?" As he elaborates, the map of his research agenda slowly reveals hidden treasures.
Unhidden AgendaEarly in their schooling, economists learn that firms grow to the size at which they minimize average production costs. This often results in large production facilities that allow companies to take advantage of economies of scale. But pushing strongly against that tendency is the cost of bringing goods to market. A single huge factory could achieve low costs per unit, but the expense of shipping products to stores across the nation could overwhelm the scale economy savings. That balance is at the heart of Holmes' research agenda. "If you pile up production in this one little spot, you can have great efficiencies, but how are you going to get it to people?" he asks. "That's really the trade off that drives much of my work: enjoying scale economies versus saving on transportation costs."
A recent Holmes paper published in the "Journal of Political Economy", for example, examines why companies tend to put sales offices in large cities. One potential explanation is that salespeople in large cities have greater access to market information that enables them to match customer needs. But by carefully constructing a model and comparing it to U.S. data, Holmes finds that scale economies and transportation costs may be more persuasive reasons.
Economics of DensitySimilar factors play a role in his Wal-Mart research, which examines the time-path of the corporation's store openings across the country. It might seem that Wal-Mart would have chosen the highest quality locations around the country, one by one, as it expanded its empire. But instead, Holmes concluded, the company pursued economies of density: locating new stores close to older ones in order to save money on deliveries, infrastructure, management and advertising. A short animated movie clip created by Holmes tracks the company's diffusion across the United States from 1962 to 2004 and shows it spreading like a virus--or pollinating like a flower, depending on your point of view.
While scale and transportation costs are critical baseline factors, they alone can't explain why industries locate where they do. Government policies play a significant role, as does labor union activity. "I often look at how avoidance of labor unions or government regulations might tilt company decisions," Holmes says. He uses his benchmark models incorporating scale and transportation to then measure the separate impacts of regulation and union activity.
Digging DeeperA unifying theme in Holmes' research is the importance of digging hard for data and comparing those data tightly with theory. Most economists lean toward either theory or econometrics. Not Holmes. "A lot of people do well in this field by just doing the math, sitting in their office with pen and paper," he observes. "And others are really specialists with data. It's just worked well for me to try to play both cards."
So while his papers are filled with high theory and intricate math, they also incorporate gigabytes of data that are meticulously assembled from government databases, industry sources, local phonebooks or online articles. And while he could outsource the data gathering to others, Holmes chooses to do much of it himself. "Yeah, I get dirty," he jokes. "I'll do a lot of the data work; at the very early stages I always need to do it."
For Holmes, a direct relationship with the data is essential. "It's never enough for me to run some regression and just see what pops up," he says. "I look at my data and look at the outliers, and I say, what's their story? You learn a lot by digging. You often learn about something new that you didn't even know existed. If I didn't dig I'd miss out on my best ideas."
Holmes' SchoolingHolmes's zeal for economics started with his first college seminar. "I went to my first econ class never even having heard the term 'supply and demand.' I had no idea what it was about," he recalls. "And, well, I just fell in love with it. I couldn't believe it. I basically never did anything else."
He majored in math and economics at the university of Pennsylvania. Then he got his Ph.D. at northwestern university in 1985. After teaching at the university of Wisconsin, he became a Federal Reserve Bank of Minneapolis economist in 1993, joining the university faculty two years later. In 2003 he was awarded the Carlson Chair. In addition to his research, Holmes teaches both graduate and undergraduate classes at the university. "I really enjoy undergraduate teaching," he says, recalling his own initial classes in the field. But working with graduate students is especially gratifying. "I get the most out of the one-on-one working with students on their theses," he says. "It's really thrilling to see somebody put their research together. One of the most rewarding aspects of this job is to see your grad students grow."
Holmes' enthusiasm about economics begs an obvious question about his own children, the eldest of whom has just started college. Would he predict that any of the three will follow in their father's footsteps? Holmes smiles widely. "That's still to be determined," he responds, always the cautious scholar. "I'm hoping. I don't want to be too pushy. But I think it's the greatest job--and the greatest field--in the world."