Fearful of the growing market power of online travel agencies (OTAs), regulators are pushing back against contracts which prevent hotels from offering rooms directly to consumers at a lower rate. As part of efforts at TSE Competition Policy and Regulation Center, we strive to ensure that policy interventions like these are in society’s best interest. In a forthcoming article in ‘The Journal of Law and Economics’, Marc Ivaldi and his coauthors Sean Ennis and Vicente Lagos examine the impact of agreements by Booking.com and Expedia to relax their restrictions on how hotels set their prices.
Why might the rise of OTAs have had mixed blessings for consumers?
These digital platforms can play an important informational role, helping consumers to search for and compare room offers. Consumers may also benefit from OTAs when they encourage competition across hotels, help new hotels to enter the market, or allow existing firms to operate on a larger scale. However, the eyewatering commissions that OTAs charge hotels – often as much as 20% of a night’s room rate – may end up increasing prices for travelers.
What are price parity clauses and why do platforms insist on including them?
The viability of OTAs is vulnerable to showrooming. This is the practice in which thrifty consumers only use an OTA to find and compare a room; they then leave the platform and buy direct from the hotel, which can offer a lower price because it does not pay the OTA commission. Price parity clauses aim to prevent showrooming by stipulating that hotels cannot use other channels to set prices below those of the OTA.
How does your research examine the impact of price parity?
We take advantage of two regulatory natural experiments that occurred in 2015 when Europe’s largest OTAs – Booking and Expedia – agreed to relax hotels’ parity obligations. First, a broad EU intervention allowed hotels to offer lower prices on other OTAs and their own direct channels, provided that discounts were part of a loyalty program and not directly advertised to the general public. Second, France and Germany went further and eliminated all parity agreements for top OTAs.
We empirically assess the impact of these interventions on online booking prices in the EU using data from multiple hotel chains. Specifically, we compare average booking sales prices on two large OTAs and on the hotels’ own online direct channels.
The main distinguishing feature of our study is that we rely on actual transaction prices, using real sales rather than posted prices. Our data also includes sales for loyalty program customers whose offered and paid prices are unobservable from a general web search. Another specific feature of our dataset is that it includes both EU and non-EU countries, providing a control set of observations that are not subject to parity regulation. This allows us to substantially improve on empirical analysis of existing theories of price setting.
What are your key takeaways?
Overall, we find that regulating parity clauses resulted in significantly cheaper direct sales in two out of three hotel segments. Following the switch to a narrower form of price parity in Europe, we find that direct sales by mid-level and luxury hotels became cheaper than OTA prices. This is confirmed by comparison with non-EU prices. However, we find opposite results for budget hotels, suggesting that different economic forces or factors may be at play.
Full parity elimination appears to have had stronger effects only for mid-level hotels in Germany. However, the primary effects that we observe come from the narrower EU-wide policy, casting doubt on the effectiveness of the Franco-German total ban on price parity. Alternatively, our results may have been influenced by hotels reducing their direct prices in both EU and non-EU countries. For instance, prices paid by EU citizens for non-EU hotels may have been indirectly influenced by the new policy.
What future directions do you envisage for research in this area?
Beyond the specific impacts of parity clauses in the hotel sector, our results also provide empirical evidence on the impacts of most-favored-nation (MFN) clauses more generally. While they are not necessarily directly applicable to other industries, our results may provide information on default expectations when key industry characteristics are similar. For example, similar questions arise with credit-card companies’ restrictions on what merchants can charge customers based on the payment mechanisms they use.
Future work could develop a structural model of supply and demand for different channels and hotel types to understand how MFN clauses affect buyer and seller reactions, particularly for different hotel segments. It would be interesting to explore how substitution patterns affect outcomes and the extent of platform market power in the hotel sector.
‘Price Parity Clauses for Hotel Room Booking: Empirical Evidence from Regulatory Change’ and other publications by Marc Ivaldi are available on his TSE web page.
- Marc Ivaldi
- Sean Ennis
- Vicente Lagos