Doctorate in Business and Socioeconomic Sciences
Research Profile Egon Smeral
Full Professor, Department of Tourism and Service Management
Description of Research Profile
I started my career 1974 as a scientist in the Austrian Institute of Economic Research (a think-thank with an excellent international reputation; the institute was founded in 1927 by Friedrich von Hayek and Oskar Morgenstern) . My first assignment at the institute was to do research into international economics, especially foreign trade. In the first years as an economic researcher I investigated potential Austrian export markets at a country level, wrote papers about the export industry, current account problems and exchange rate effects on foreign trade, and started to model merchandise exports and imports at a national and international level (Smeral, 1979). I was also involved in econometric forecasting and economic policy advising. From the very start I had to deal with empirical facts and data; only the study of theories was not part of my job description. This working philosophy has not changed and will not change.
In the eighties I discovered “tourism” as an interesting research field. First, in several publications I brought together consumer and foreign trade theory to explain tourism demand. This development process was started in 1988 and finished by 2004 (the cornerstone publications are: Smeral, 1988, 1994, 2000, 2004). Here is a short outline of the model:
The model can be viewed as a static partial model and explains country-specific tourism imports and exports based on payment flows at constant prices and exchange rates. Static partial models start out from a multi-stage budgeting process. Multi-stage decision-making and budgeting means that overall expenditures are first broken down by time, and then followed by the formation of budget groups, before the group budget is assigned to individual items. In other words, after allocating the budget by time, it is next allocated by leisure-time goods and other consumer goods. After that, the budget allocated to international travel is separated from the budget for domestic traveling and the budget for other leisure-time goods. In a last stage, a distinction is made between destination countries (the expenditure streams are tourism imports of the origin country and tourism exports of the destination country). The model explains tourism imports by an income effect, a price effect and special factors. Tourism exports in turn are explained by a demand and price effect, plus the effects of special factors. Key variables are the relative prices in a unified currency (tourism-weighted real effective exchange rates) and GDP variables as an indicator of income variables. In order to incorporate the range of other important explaining variables, dummy variables and specific trend variables have to be used. Each country included in the model is both a country of origin and destination. The (endogenous) variables explained by the model are the respective tourism exports and imports at constant prices and exchange rates.
Drawing on this type of model outlined above I prepared and published several long term forecasting and impact studies. The impact analyses were mostly the outcomes of historical and political events (e.g. the fall of the Iron Curtain, German re-unification, several EU enlargements, introduction of the euro, etc.). In addition to causal econometrics I used univariate time series econometrics for forecasting, measuring forecasting accuracy and impact analysis such as the effects of the EU presidency (e.g. intervention models; Smeral – Wüger 2000 and 2008; Smeral – Wüger, 2005 and 2006; Smeral, 1993; Smeral-Weber, 2000; Smeral, 2007a). Forecasting has always been an important activity in my research work (and will continue to be so in the future). This can be seen from a technical/methodological side and a policy adviser side (I have given tourism policy advice to the Austrian Ministry of Economic Affairs and the National Tourism Board for about 20 years).
Around 12 years ago, Peter Laimer from Statistik Austria and I developed a Tourism Satellite Account (TSA) for Austria, later followed by regional Satellite Accounts for three Austrian states. For these accounts we do an update every year. In the case of the TSA-topic I pointed out critically that the TSA spans only those effects that are generated by the direct economic relationship between guest and producer and thus makes it difficult to compare tourism-related GDP in relation to the overall GDP, since the latter also includes indirect effects caused by economic interlinkages (Smeral 2005 and 2006). Further, I showed that a problem arises from the fact that, in the TSA, expenditures from residents on business trips are accounted for as final demand. On the other hand, intermediate consumption is not considered in GDP calculation, resulting in a biased comparison of the value added to GDP according to the TSA. It is clear that in measuring the TSA-based contribution (the „impact“) made by the tourism industry to national/regional GDP, results must be adjusted for indirect effects and intermediate consumption (as we did for Austria at the national level and for the three Austrian states). The adjustments demonstrate the enormous scope by which the tourism industry's contribution to national and regional GDP is underestimated, even when only its direct effects are considered, as is required by the TSA approach. Even when business trips by residents are included and thus make for a greater contribution of tourism to GDP, this fails to compensate for the downward bias produced by ignoring the indirect effects of tourism demand.
Looking back, I have to say that impact studies have kept me very busy for the last three decades and I see no end. Douglas Frechtling seems to have a similar history so that we decided to do a little survey together about impact measurement (little because the publisher did not allow more than 5000 words; Frechtling-Smeral, 2010).
Growth theory was another research field which attracted my interest: impressed by the work of Robert Solow and his intention to draw a parable of economic growth, I tried to formulate a parable of tourism growth (Smeral, 2001, 2003, 2007b and 2009a; Solow, 1970). For this research project I focused on structural aspects because these are important factors when it comes to explaining tourism growth (the Vienna School of Economics is influenced by structural thinking, which certainly had at least some influence on my scientific work). In this connection, crucial roles are played by the structural change in demand and differentials between productivity in tourism and in manufacturing.
Another core area of my research work is the business cycle and the connected fluctuations in the consumer behavior. In dealing with the impacts of global financial and economic crisis I found myself increasingly doubtful about whether (tourism) demand elasticity is really symmetrical, i.e. that the elasticities in up- and downturns have a similar magnitude or that they remain stable across the business cycles (Smeral, 2009b, 2010 and 2011). On a country level I found clear empirical evidence that the relative fall in (tourism) demand during an economic downturn is more precipitous than the relative increase in demand during an economic upturn of a similar magnitude. In two papers I challenged the mainstream literature assuming symmetric income and price effects as a standard. Using a modified growth rate model I showed that tourists’ demand reactions are asymmetric in different phases of the business cycle (Smeral, 2012a and b). Specifically, this is to say that tourists will behave differently when the economy grows in terms of the real GDP faster than the flexible trend as compared with that situation when the economic growth is below the growth rate of the flexible trend. Main reasons for varying income and price effects are "loss aversion" in consumption standards and liquidity constraints. Additionally, the existence of precautionary saving, changing behaviour of households to get into debt or to save as an outcome of habit modification, supply factors such as financial innovations as well as the intensity and time structure of substitutions between expenditures on outbound travel, domestic stays and other goods and services play also a role.
In a paper together with Haiyan Song we compared the forecasting performance of the modified growth rate model from above with the results of the time varying parameter (TVP) method (Smeral-Song, 2012). The results show that the TVP model does not perform well as suggested by the previous studies and that the modified growth rate model delivers clear the better results. The latter is especially true for periods with strong business cycle fluctuations such as 2009 and 2010.hh
To summarize, the overview given above shows competences in the research fields of
• Quantitative Economic Research in General as Well as Economic Policy Advising
• Econometric Modelling and Forecasting
• Impact and Growth Analysis
• Business Cycle Analysis (Especially with Respect to Consumer and Tourist Behavior)
Selected Theses Supervised
1. Valeria Croce, Expert's predictive power - an explorative study on the Aggregation and integration of judgmental and statistical forecast, 2011/2012 ongoing