The research literature on Wal-Mart: Some frowns, some smiley faces
Hoping for some easy answers to the Wal-Mart effect? Better shop elsewhere.
Published January 1, 2008 | January 2008 issue
Someone unfamiliar with the Wal-Mart controversy might reasonably assume that very obvious macro effects sprout to life when Wal-Mart opens its doors in town. How else could one rationalize the ferocity, the righteousness of the opinion that surrounds the firm?
In fact, it's the exact opposite; much of the debate and controversy likely get their fervor from the fact that nobody knows—for sure—who's right about Wal-Mart's effect on local communities.
Despite a considerable amount of research, a definitive consensus about Wal-Mart's effect on local communities is hard to declare. And it's not for lack of effort. Arguably, no firm in the history of commerce has been more scrutinized. Business schools routinely analyze successful firms to discover their secrets of success. But Wal-Mart is one of the few businesses routinely, even energetically, put under the microscope to look for its dark side, to determine whether its genes are compatible with a community's DNA.
That's not to say there are no conclusions whatsoever from the research. In fact, there are conclusions both damning and supportive of Wal-Mart. This is the likely source of the debate's din: When there is doubt or disagreement, talk louder.
Similar to the results reached by this fedgazette analysis, empirical research on the economic effects of Wal-Mart tend to be modest in both directions. The balance of academic research probably favors Wal-Mart on the whole, but not by a lot, and not in all measures.
What follows is a summary of some of the major and recent work done on the local economic effects of Wal-Mart on measures similarly used in this fedgazette analysis—jobs, firms, wages and poverty. One additional measure—beyond the scope of our study but very relevant to the debate over local effects—is also discussed here: consumer welfare.
Jobs: Wal-Mart's effect on employment is instructive of much of the debate. In a widely cited study (1) first published in 2004, Emek Basker found that about 100 jobs were gained in small retail over the short term, falling to 50 in the long run as smaller employers folded or contracted. Outside of retail, another 20 jobs were lost in wholesale trade, thanks to Wal-Mart's vertical integration. A study (2) one year later by Michael J. Hicks showed a similar employment uptick of about 50 jobs in Pennsylvania counties where Wal-Mart opened new stores in 2002. His study also found that Wal-Mart's entrance was associated with a significant decline in retail job turnover.
Researchers David Neumark, Junfu Zhang and Stephen Ciccarella disagreed with those results in a 2005 study (3) published by the National Bureau of Economic Research (NBER). The trio found that a Wal-Mart store opening reduced county-level retail employment by about 150 workers, with each Wal-Mart worker replacing approximately 1.4 retail workers and lowering average retail employment in a county by 2.7 percent.
Then a return volley came a year later in a working paper (4) by Scott Drewianka and Dain Johnson, which found that Wal-Mart raised local retail and possibly even nonretail employment. Still other studies have pointed out that Wal-Mart likely induces greater efficiency and productivity in surviving firms, meaning that although employment among such firms might decline because of Wal-Mart, they might also be more competitive than in pre-Wal-Mart days.
The fedgazette analysis found slightly positive employment effects in counties with a Wal-Mart. About the most that can be said about Wal-Mart's effect on jobs is that it is small—even by the standards of counties with modest populations—which itself might be a useful point, given the current rhetoric on both sides.
Firms: Much research attention is also given to Wal-Mart's supposed demolition of existing businesses in a community. The evidence suggests that, indeed, retail competitors suffer when Wal-Mart enters the local market.
In 2005, Panle Jia (5), looked at the effect of Wal-Mart and Kmart on discount retailers and found that entry by either store made about 50 percent of discount stores "either unprofitable or unable to recover their sunk cost." That sounds like a lot, but represented only two or three firms per county because of the study's narrow focus on discount firms. Basker's 2004 study widened the lens to look at all small retail establishments and found a closure rate of about 2 percent (or about four of 200 firms on average).
Few studies look at the effect on establishments outside the retail sector. Though stories of business closures are rife, anecdotal evidence is also gathering of new firms capitalizing on Wal-Mart's ability to bring traffic to town. Two studies completed in 2006 suggested that Wal-Mart might have a positive influence on firm creation within a county. Drewianka and Johnson found a small but consistently positive effect on the number of establishments. "Few of the estimates are either large or statistically significant, but it is striking that the pattern is so pervasive." (The fedgazette analysis found a similarly small, but persistently positive, effect on the number of firms.)
Russell S. Sobel and Andrea M. Dean (6) examined the rate of self-employment, the number of small employer establishments and the profitability of small businesses using both time series and cross-sectional data. Their work found "no statistically significant long-run impact on the overall size and profitability of the small business sector in the United States."
Sobel and Dean acknowledged, and then rebutted, the common notion that long-established, locally owned businesses get replaced by "lesser" small businesses. But they pointed out that both revenue and net income for small businesses have continued to rise. They wrote, "While the entry of a specific Wal-Mart store might cause some individual small, 'mom and pop' businesses to fail, our results suggest that these failures are completely offset by the entry of other new small businesses somewhere else in the economy."
But even for surviving firms, Wal-Mart's presence is believed to cut into revenues. A 2002 study (7) by Kenneth E. Stone, Georgeanne Artz and Albert Myles looked at sales trends in Mississippi counties with and without a Wal-Mart supercenter. It found that total sales in host counties increased by small but notable amounts, while nonhost counties experienced a decline of similar degree. Some individual sales categories, like general merchandise, saw much larger shifts.
Wages: To borrow one of Wal-Mart's own advertising phrases, the company is widely perceived to always have the lowest wages and is regularly accused of pulling down overall wages. There's not much evidence of either, in part because the matter has not been studied very thoroughly, according to some.
A 2007 working paper (8) from the Institute for Research on Labor and Employment, housed at the University of California, Berkeley, pointed out that to date "there is little academic work on how the company's expansion may have transformed the retail wage structure."
Its authors, Arindrajit Dube, T. William Lester and Barry Eidlin, looked into the matter and concluded that one new Wal-Mart reduced retail earnings per worker by 0.5 percent at the county level, and 10 new Wal-Marts reduced earnings between 0.5 percent and 2 percent at the state level, most of it in retail subsectors and grocery. However, the wage effect was evident only in metro counties. Nonmetro counties showed no evidence of wage reduction, which the authors argued was likely the result of lower average wages in general in rural areas. (The fedgazette study looked exclusively at nonmetro counties and found small but positive effects on retail earnings per job.)
Neumark et al. found that Wal-Mart's presence leads to a decline in a county's retail earnings of 1.3 percent, but they pointed out that there is no evidence of reduced retail earnings per worker. Drewianka and Johnson found that supercenters were associated with a 1 percent decrease in wages.
But that's not the final word. Hicks (2005) found no effect on existing employee wages in the retail sector. However, new hires in the retail sector saw a roughly 50-cents-an-hour increase in total compensation in the quarter Wal-Mart entered, coupled with a finding showing that retail employee turnover also declined with Wal-Mart's entry. A 1999 study (9) by Hicks and Kristy Wilburn also uncovered an increase in retail wages in 14 West Virginia counties studied.
A major gap in the research regarding wages has to do with workers' previous wages and work intensity (full- or part-time) compared with Wal-Mart jobs. Critics regularly rail at the company for paying low wages. But it's hard to fathom how the company could attract the necessary labor—250 to 400 workers for supercenters—if compensation paid by the retailer was not on par with job opportunities elsewhere, including applicants' existing jobs, where relevant. This is particularly so given little evidence of wide-scale loss of either competing jobs or firms when Wal-Mart enters the local market.
However, as Basker pointed out in a 2007 study (10), there is some anecdotal evidence that Wal-Mart can push wages down where the retail sector is unionized or otherwise highly compensated—mostly in larger urban markets—because firms sometimes have to seek concessions from higher-paid employees to compete with Wal-Mart's lower cost structure. But the retail sector is also one of the least unionized and lowest-paying sectors of the economy, which means these wage confrontations may be more exception than rule.
Poverty: Despite the repeated attention paid to Wal-Mart and wages, very little focus has been placed on the related issue of poverty. Virtually the only study in the research literature is a 2004 paper (13) by Stephan J. Goetz and Hema Swaminathan. The authors looked at counties with Wal-Mart stores and concluded that counties with more initial stores (in 1987) and those with more store additions (through 1998) experienced greater increases (or smaller decreases) in family poverty rates over this period. (Findings from the fedgazette analysis were similar regarding poverty trends in the Ninth District.)
Consumer welfare: The one area of research with unequivocal findings is Wal-Mart's effect on consumer prices. In general, Wal-Mart prices—not on all individual items, but on baskets of common items—are substantially lower than competitors' prices. The firm also tends to induce smaller, but notable, price declines at competing stores. That translates into enhanced welfare for shoppers. (One note: The definition of "competitor" is not uniform across studies, which leads to different comparison groups; most often, price studies focus on grocery or discount stores.)
A 2003 investigation by Retail Forward, a private international retail consultant, found that prices at Wal-Mart were on average about 15 percent lower than the competition, in part because of its buyer power and lower wage scale. Two additional studies (11) by Basker (as author or co-author) since 2005 used different methodologies and found consistently lower prices at Wal-Mart (upward of 10 percent), as well as slightly lower prices at competitors.
A 2005 NBER working paper (12) by Jerry Hausman and Ephraim Leibtag found that prices at Wal-Mart were 15 percent to 25 percent lower than at traditional supermarkets. That turns into real savings. Average households saw their annual welfare increase about $780 (within the range of annual savings estimated by Wal-Mart itself, of between $500 and $1,200). The poorest households saw their welfare increase by a lower amount ($530) according to the authors, mostly because they spent less on food overall. But the savings had an outsized positive effect on poor households because their budgets were much smaller to begin with. "We find the benefits to be substantial. ... Low income households benefit the most."
Check out: The final results
So, where does all of this leave the argument about Wal-Mart's effect on local communities? Mostly right back where we started, with both sides anchored, in part because of comparatively thin evidence supporting and refuting various positions.
Dube et al. point out that few studies directly measured the effect of Wal-Mart (or big-box stores generally) on employment, wages and working conditions in the retail sector. Studies on these matters to date "produce ambiguous results and have many limitations."
And Basker added in her 2007 report, "Wal-Mart is the largest company in the world, yet little is known about its economic impact."
Maybe this tug-of-war that's going (mostly) nowhere is itself representative of Wal-Mart's full effect: The firm's positive and negative effects are mostly a wash and, at the end of the day, Wal-Mart is just a successful business, neither benevolent enough to put on a pedestal nor evil enough to shame to the economic margins.
Drewianka and Johnson's study concluded that Wal-Mart is "a fairly benign force," and they admitted to some perplexity over the debate itself. Their findings were generally favorable to Wal-Mart, but they uncovered unfavorable results as well. But they stressed that "the great majority" of estimated effects were tiny—a fraction of 1 percent in most cases. To which they concluded: "[E]stimates of this magnitude do not seem to justify the rancor of the debate. Perhaps it can be justified on other grounds, but if so it would seem wise to redirect the discussion in that direction."
1. Job Creation or Destruction? Labor-Market Effects of Wal-Mart Expansion
Emek Basker, University of Missouri, January 2004
2. What Do Quarterly Workforce Dynamics Tell Us About Wal-Mart? Evidence from New Stores in Pennsylvania
Michael J. Hicks, 2005
Air Force Institute of Technology and Marshall University
6. Has Wal-Mart Buried Mom and Pop?: The Impact of Wal-Mart on Self Employment and Small Establishments in the United States
Russell S. Sobel, Andrea M. Dean
7. The Economic Impact Of Wal-Mart Supercenters On Existing Businesses In Mississippi
Kenneth E. Stone, Georgeanne Artz, Albert Myles, 2002
8. Firm Entry and Wages: Impact of Wal-Mart Growth on Earnings
Throughout the Retail Sector
Arindrajit Dube, T. William Lester, Barry Eidlin, 2007
9. The Locational Impact of Wal-Mart Entrance: A Panel Study of the Retail Trade Sector in West Virginia
Michael J. Hicks, Kristy Wilburn, 1999
11.Selling a Cheaper Mousetrap: Wal-Mart's Effect on Retail Prices
Emek Basker, 2005
The Evolving Food Chain: Competitive Effects of Wal-Mart's Entry Into The Supermarket Industry
Michael Noel, Emek Basker, 2007
12. Consumer Benefits From Increased Competition in Shopping Outlets: Measuring the Effect of Wal-Mart
Jerry Hausman, Ephraim Leibtag, 2005
See also related fedgazette article: Thomas J. Holmes on Wal-Mart's location strategy, March 2006. Holmes describes Wal-Mart's location strategy and possible implications for the Ninth District.