Over at Anti-dismal, Paul Walker notices an article by Andrea M. Dean and Russell S. Sobel which show that the regions with a Wal-Mart tend to have a higher number of small enterprises (with 5-9 employees) than regions without a Wal-Mart. The same trend holds (albeit more weakly) for businesses with 1-4 employees, and the trend is flat for self-employment.
This is an interesting result, given that the majority of the public tends to believe that the existence of stores such as Wal-Mart and the Warehouse have driven mum and dad retailers out of business. Now we could argue about whether having large stores drive out small stores is a good thing or not (I think it generally is), however this article suggests that there may not even be a trade-off, in which case society should stop hating on these large firms for “destroying mum and dad stores”.
As a result, lets ask ourselves what the article’s cross-sectional data means.
What about co-movement?
So, could some external factor be driving these differences. For example, if a certain region is wealthier, it may be the case that Wal-Mart is more likely to enter the market and also likely that we will have more small stores. Furthermore, if one region has a higher population, it can sustain more stores and a Wal-Mart, while a region with a small population can’t. Also, if the region is larger, it will require a larger number of stores.
Now, the paper actually includes all these factors in their analysis – so the trend they find is independent of this. The only potential problem we could run into endogeneity – where the co-movement is occurring because of a variable that they have not included in their model (or because the dependent variables depend on the number of small businesses in some way). However, they also correct for this.
Furthermore, in this case they find that the 1-4 employee businesses are the ones that grow (statistically significantly) when a Wal-Mart is around.
As a result, it appears to be the case that Wal-Mart either has little impact on the number of small businesses in a region, or actually increases the number, how could this work?
Well the two explanations I can think of are:
- Small businesses may provide services/products that complement Wal-Mart,
- A Wal-Mart provides goods at a lower cost than previously – implying that labour productivity would be higher (as the value-added per employee is higher). This should lead to an increase in economic activity and incomes – which would feed into other industries in the region.
You say you still don’t like it!
Ok, if you still want to hate on the large corporations you can say something like:
Although the number of small firms might not change, the type of small firms in equilibrium could well change (eg the mom and pop dairy could be screwed), which could cause a loss of welfare (through some “community spirit” mechanism, or the belief that welfare of the traditional stores is more important than the new ones).
I wouldn’t agree with this – but it is still “valid”, and it would be transparent.
However, this data suggests those people who say that they don’t like Wal-Mart and the Warehouse because it destroys small communities will need to do a bit better then saying it drives small businesses out!