Distribution, convenience, and market institutions


Convenience means ease to obtain or reach. Substantial amounts of resources are being used to produce convenience to consumers. From the National Accounts Data Base at the United Nations, the expenditure in a typical country on retailing and wholesaling is close to 22% of GDP, irrespective of country wealth, ease of doing business, proxies for consumer travel costs, or proxies for the modernity of the national retail system.

The production of convenience adds value. Below is graph of the distribution across nations of the national value added by retailing vs  the national value added by manufacturing.  Both sectors add about the same amount of value. Like making products, making products conveniently available is important economic activity.

My interest in convenience is based in part on the idea that using the market to buy products takes effort on behalf of the consumer and that the production of convenience can be viewed as a transfer of consumer cost of purchasing to firm cost of selling. More specifically, I am interested in how competitive forces divide the total cost of using the market into a firm cost of selling and a residual consumer effort of buying.

Convenience also means ease of use. I am currently working on projects that investigate how time availability impacts variety and the nature of consumption goods bought by consumers.

Value added by retailing vs. manufacturing


NOTE –  Panel (A): all nations in the UN National Accounts; Panel (B): nations with above median GDP. The value added by retailing in most wealthy countries is close to the value added by manufacturing.  Source: “The Provision of Convenience and Variety by the Market,” RAND Journal of Economics, 2015, 46:3 (Fall), 480-498 

Brand preferences, order of entry effects


The logo to the left belongs to my favorite brand of beer, brewed in Wylre in the Dutch province where I grew up, Limburg. Named after Frederik Edmond Brand, who took ownership of the brewery in the late 1800’s, it is the most preferred beer in its local market. Different Dutch brands experience leadership in other regional markets, Grolsch in the east, Bavaria in Brabant, and so on. Interestingly, these brands have been able to successfully defend their historical position head-to-head against large opponents such as Heineken.

Local market share leadership and sustenance of local advantages are very common in European and United States markets. Manufacturers of branded goods often have a large market share premium in markets where they were launched early and enjoy strong consumer loyalty towards their brand, even when the products they sell are close substitutes. With several co-authors, I am working on answering questions about the origins and the persistence of these effects,  and about the underpinnings of brand preferences.

Order of entry and market share


NOTE – The joint geographic distribution of share levels and early entry across U.S.markets in ground coffee. The areas of the circles are proportional to share levels. Shaded circles indicate that a brand locally moved first. Source: Journal of Political Economy, 2009, 117:1, p. 99.

Consumer search, online markets


Product search is costly, in terms of effort and time. This cost limits the set of products consumers consider for purchase.  With co-authors, I study how consumers search for durable goods online. I am interested in what happens to search and choice behavior if search becomes less costly or if information about products is organized differently.

Using large consumer panels, we find that consumers often search products that are quite similar to the ones they buy (see the figure below for a recent example using comScore’s panel data). Related to this stream of work, together with colleagues, we are working on developing methods that predict what consumers buy, using data covering what they search. This is especially useful in estimating demand for consumer durable goods (cell phones, cameras, computers, etc.), for which a market analyst usually does not observe many choices per consumer.

Search versus purchase in digital cameras


NOTE – Each plot symbol represents a consumer in the comScore panel. The horizontal axis displays the average level searched by a given consumer and the vertical axis the attribute level that was purchased. Search for  cameras is informative about purchasing behavior and demand. Source: “Zooming In on Choice: How Do Consumers Search for Cameras Online, ” Marketing Science (forthcoming)with Jun B. Kim, and Carl F. Mela.