Passar para o conteúdo principal

Operations & Data Science

Determinants and Heterogeneity of Switching Costs in IT Outsourcing: Estimates from Firm-Level Data

Christian Peukert

Firms across a wide range of sectors increasingly rely on the outsourcing of information technology (IT) services, mostly to save on costs and therefore enhance productivity. Because not all firms are equal, of course, outsourced IT services are usually not completely standardized and require customization. Hence, IT services cannot be bought on a spot market, but require the negotiation of individual contracts, often involving just a few parties. A long tradition of research suggests that contracts are doomed to be incomplete, because nobody can perfectly foresee all contingencies, which in some cases can invite opportunistic behavior of the involved parties.

In this paper, I study how firms decide whether to continue an existing relationship or switch the vendor of outsourced services. It has been widely acknowledged that there might be potentially high costs associated with switching. Cases studied in the literature suggest that switching costs are “large”, “high”, “extremely high”, “massive” or “the single most threatening aspect of IT outsourcing”. However, precisely measuring switching costs has been difficult for a number of practical reasons. One of those reasons is that these costs are hidden, very much like the well-known concept of opportunity cost, and not reported in the internal or external accounting of firms. The methodological contribution of my paper is that I develop a statistical model to deal with such issues.

In the context of US community banks that outsource information technology, my estimates suggest that average switching costs account for about 30% of the average annual expenditure on professional services. For some firms the switching costs are even higher. In fact, high enough to prevent them from switching after all. I estimate that this so-called "lock-in effect" is equivalent to about 70%.

A second contribution of this paper is the decomposition of switching costs into market-specific, firm-specific and relationship-specific determinants. Our results show that the largest portion of switching costs is due to relationship- and firm-specific factors, while market-specific variables are much less important.

However, this is not all bad news. My results also imply that firms can have substantial discretion in affecting the size of switching costs. In fact, about 80 percent of switching costs can at least partially be affected by the firm. I show that firms can lower switching costs disproportionately by investing in vendor search and evaluation. On top of that, experience effects and scale economies reduce switching costs. The largest portion of switching costs is driven by relationship-specific variables such as relationship duration.

The paper is therefore not only a statistical measurement exercise, but the findings provide important implications for the management of switching costs, both from a firm- and vendor-perspective.

Peukert, C. (2018). Determinants and Heterogeneity of Switching Costs in IT Outsourcing: Estimates from Firm-Level Data. European Journal of Information Systems.

Demand variation, strategic flexibility and market entry: Evidence from the U.S. airline industry

Christian Peukert

Changing demand levels are a key strategic challenge for firms, as they must adapt their capacity either downward to maintain high levels of resource utilization or upward to avoid missing earning opportunities due to capacity constraints. Reasons for demand variation are manifold and can range from predictable seasonal changes in consumption to unexpected macroeconomic shocks.

In this paper, we study a strategic safeguard against market turbulence: We ask if, and under which circumstances, flexible production processes can help firms to enter markets that are characterized by fluctuations in demand.

We develop hypotheses regarding the role of demand variation and flexibility in market entry decisions. Specifically, we hypothesize that demand variation leads to less market entry, that flexibility is associated with a higher probability of entry in new markets, and that this relationship is stronger with higher levels of demand variation. We go on to argue that the benefits of flexibility are greater when variation in demand is less predictable, further increasing the likelihood of entry.

We take these hypotheses to the data in the context of the airline industry.

In this industry, markets can be clearly defined by city pairs and entry happens frequently. We observe the entry decisions of 75 airlines into 2,065 domestic markets in a sample of more than 11 million observations between 1995 and 2013. Industry institutions are standardized across markets and stable over time. We measure flexibility as the number of different aircraft types that are at the disposal of an airline.

Our results show that airlines that choose strategic flexibility over operational efficiency are more likely to enter volatile markets, especially when this variation is difficult to predict.

Given that flexibility is a strategic choice that comes at a cost, our study has implications that are of vital practical importance. We argue that flexibility requires organizational and operational structures that are not easily created. Managers must weigh the benefits of efficiency against the merits of flexibility. In settings where it is difficult to predict future demand with historical data, our findings may also imply that strategic flexibility can complement investment in business analytics, helping managers to react to environmental change that is difficult to foresee.

Claussen, J., Essling, C. & Peukert, C. (2018). Demand Variation, Strategic Flexibility and Market Entry: Evidence from the U.S. Airline Industry. Strategic Management Journal.

Catch Me if You Can: Effectiveness and Consequences of Online Copyright Enforcement

Christian Peukert

We evaluate the takedown of, a major platform for unlicensed video streaming in the German market. We compare how the web behavior of a large number of users in Germany has changed compared to users in three other European countries. We find that the intervention was not very effective in reducing unlicensed consumption or encouraging licensed consumption, mainly because users quickly switch to alternative unlicensed sites. To the contrary, we find that press coverage of the takedown operation allows consumers to learn about the existence and legal status of unlicensed video streaming, causing some users to become pirates.

Our findings may perhaps translate into other contexts, where aggregate behaviour has seemingly not changed quite as intended, despite law enforcement efforts, e.g. in the so-called war on drugs. In such cases, the design of alternative policies seems to need to be informed by empirical evidence of the underlying mechanisms. Managers in the movie industry, as well as legislative bodies that are confronted with the challenges of digitization and the design of private and public law enforcement policy, are most likely to be impacted by our research.

Peukert, C., Aguiar, L. & Claussen, J. (2018). Catch Me if You Can: Effectiveness and Consequences of Online Copyright Enforcement. Information Systems Research.