Quantifying and Accounting for Quality Differences in Services in International Price Comparisons: A Bilateral Price Comparison between United States and Japan

Purchasing power parities (PPPs) from the International Comparison Program (ICP) are used for cross-country comparisons of price levels and real gross domestic product (GDP), household consumption and investment. PPPs from the ICP are also used in compiling internationally comparable output aggregates and making productivity comparisons in the KLEMS initiative. PPP compilation is anchored on the principle of comparing the like with like and price data are collected for goods and services with detailed specifications in the form of structural product descriptions. While this approach works well for goods, it is not effective in the case of services. If differences in service quality exist, these get reflected in the PPPs from the ICP. In this paper, we focus on the USA-Japan bilateral price comparison in the 2014 ICP in the OECD region and estimate bias induced by differences in quality of services in. Service quality is driven by various unobservable factors. In this paper we make use of data on quality differences and consumers’ willingness to pay collected through a specialised survey conducted by the Japan Productivity Center early in 2017. Data are collected from a large sample of 517 respondents from USA and 519 respondents from Japan, covering 28 service items including transport, restaurants, retail services, health and education. Estimates of consumers’ willingness to pay for quality differences in services by the US and Japanese consumers are obtained using standard econometric methodology, these are in turn used in estimating quality adjustment factors that can be applied to price data used in PPP computation. Using the Sato-Vartia index, which has useful analytical and decomposition properties, we find PPP for household consumption (including real estate services) of 113 JPY per US dollar reduces to 104 JPY per dollar after adjusting for quality differences. When real estate services are not included, PPP reduces from 95 JPY to 87 JPY after quality adjustment. The paper also presents labour productivity estimates before and after quality adjustment for a number of service sectors including transport and storage; retail trade; hotels and restaurants; and other subsectors. Our exploratory study demonstrates that adjustment for quality differences in services is feasible and such adjustments are important for making meaningful international price comparisons.44 p