A MEASURE FOR REGIONAL RESILIENCE TO ECONOMIC CRISIS

Markowska, Małgorzata Article A MEASURE FOR REGIONAL RESILIENCE TO ECONOMIC CRISIS Statistics in Transition New Series Provided in Cooperation with: Polish Statistical Association Suggested Citation: Markowska, Małgorzata (2015) : A MEASURE FOR REGIONAL RESILIENCE TO ECONOMIC CRISIS, Statistics in Transition New Series, ISSN 2450-0291, Exeley, New York, NY, Vol. 16, Iss. 2, pp. 293-308, https://doi.org/10.21307/stattrans-2015-016 This Version is available at: https://hdl.handle.net/10419/207774 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ STATISTICS IN TRANSITION new series, Summer 2015 293 STATISTICS IN TRANSITION new series, Summer 2015 Vol. 16, No. 2, pp. 293–308 A MEASURE FOR REGIONAL RESILIENCE TO ECONOMIC CRISIS 1 Małgorzata Markowska 2 ABSTRACT The purpose of the study (presented in this article) was to develop a measure of resilience to crisis, one that may be applied to regional data. In principle, such measure can take either positive or negative values. A positive value confirms resilience to crisis, whereas a negative one confirms the absence of resilience (sensitivity/vulnerability). The measure uses growth rates referred to the previous year under the assumption that crisis results in a slowdown in growth, or even in a decline in values of important economic indicators. Growth rates are standardized by dividing values of original change rates by medians specified based on spatiotemporal data modules. Such division results in each characteristic being brought to equal validity. Simultaneously, the original character is maintained and variables are not “flattened” by the outliers. Changing destimulants into stimulants occurs during growth rates calculation. The measure of resilience to crisis is calculated as an arithmetic mean of the values of characteristics brought to comparability. The measure of resilience can be converted into the measure of sensitivity by multiplying it by (-1). The application of the proposed measure to assessing the resilience to crisis in the period 2006-2011 is presented for regions meant as the European Union NUTS2 units. The measure is based on comparable data, which allowed for using only six variables measuring changes in GDP, salaries, investments, household income, employment and unemployment. Key words: economic crisis, aggregate measure, NUTS 2. 1. Introduction Economic resilience to crisis with reference to a region is defined as its economic capacity to overcome negative external impacts. It depends on macroeconomic factors and internal determinants. Among macroeconomic factors the following can be listed: fiscal, economic and monetary policy. Internal factors 1 The project has been financed by the Polish National Science Centre, decision DEC- 2013/09/B/HS4/00509. 2 Wrocław University of Economics. 294 M. Markowska: A measure for regional… take the form of, e.g.: economic structure, restructuring and modernization level of enterprises, competitiveness and innovation. Among the important internal factors the level of human capital, including entrepreneurship, is also considered (Masik, Rzyski 2014). The objective of the article is to present the proposal for the construction of a measure of resilience to economic crisis, possible to be applied to regional data. 2. Sensitivity to crisis – research overeview The assessments of economic reactions to shocks resulting from, e.g. an economic crisis are performed by analyzing macroeconomic sensitivity specified: - in a more extensive sense as the “vulnerability to external factors distracting a particular economy from following the desirable trajectory of development” (Zaucha et al. 2014: 208), - whereas in a narrower sense (sensitivity) in the context of “economic structures and their tools for weakening negative stimuli and threats, as well as deriving benefits from the occurring opportunities without any structural changes” (Zaucha et al. 2014: 208). The studies of resilience and sensitivity to macroeconomic impacts, covering especially small countries, have been conducted for twenty years both independently and in a team by L. Briguglio (Briguglio 1995) from the University of Malta. The team’s output includes, among other things: methods for the “construction” of economic resilience in small countries (Briguglio, Kisanga 2004, Briguglio, Cordina, Kisanga 2006, Briguglio 2014), developing the concept and measuring both sensitivity and resilience (Briguglio et al. 2006a, Briguglio et al. 2009), updating and extending the Economic Vulnerability Index (Briguglio, Galea 2003), the proposal of sensitivity and resilience profiles (Briguglio et al. 2010), the identification of economic resilience pillars in small countries (Briguglio et al. 2008), the analysis of growth problems in terms of resilience (Briguglio, Piccinino 2012), the assessment of economic resilience and adaptation potential (Briguglio, Cordina 2003). Moreover, the studies focused on regional resilience were carried out by, e.g.: S. Christopherson, J. Michie and P. Tyler (2010) – theoretical and empirical aspects, K. Chapple, M. Belzer (2010) – job market, G. Bristow (2010) - competitiveness, J. Clark, H.-I. Huang and J. Walsh (2010) – innovation districts; R. Hassink (2010) as well as A. Pike, S. Dawley and J. Tomaney (2010) – differences in regional adaptation. The research team, under the leadership of P. Churski (The National Centre for Science Project entitled: Socio-economic growth vs. the development of growth and economic stagnation areas (2011-2013)) conducted research the results of which STATISTICS IN TRANSITION new series, Summer 2015 295 are available on the project website: www.owsg.pl. The identification and assessment cover growth and stagnation areas based on the set of 49 indicators divided into five blocks (population and settlement, job market and economy structure, technical infrastructure and spatial availability, financial situation and wealth level, innovative economy and business environment), whereas within the framework of blocks – the factors described by means of qualities characteristic for a given factor (Perdał, Hauke 2014: 71). The studies presented by the research team are mainly focused on the territory of Poland (various NUTS levels), with particular emphasis on Wielkopolska region. For the purposes of performing comparisons the data from Slovakia, Lithuania and Latvia were used, among other things. The identification of factors and analyses were carried out with reference to the following groups of spatial units: all units, growth areas, transition areas and stagnation areas, mainly in the period 2000-2010. The research on resilience to crisis, especially in Pomorskie region, is conducted within the framework of the project: Economic Crisis, Resilience of Regions – ESPON 2013 (partners: Cardiff University (project leader), FTZ- Leipzig, Aristotle University, Tartu University, University of Gdańsk, Manchester University, Experian Plc.), the purpose of which is (Masik 2013): “the identification of economic crisis impacts on regional economies, the analysis of structural and functional determinants in regions, an attempt to answer the question why some regions are more resilient than others, the identification of policies supporting economic resilience”. The team under the leadership of J. Szlachta (Zaucha et al. 2014: 206-234) conducted the review of the subject literature in terms of approaches to regional sensitivity measurement within the framework of the project – The sensitivity of Polish regions to challenges of contemporary economy. Implications for regional development policy, grant from the National Centre for Science 1635/B/H03/2011/40 and within the framework of project implementation supervised by D. Strahl entitled: “Smart growth vs. sensitivity to economic crisis in regional dimension – measurement methods” (grant from the National Centre for Science 2013/09/B/HS4/00509) M. Markowska (2014), focused on such areas as: economy, job market and households, listed as the most vulnerable in the context of crisis phenomena assessment. 3. Proposal for measuring regional resilience to economic crisis (RRC) It has been initially assumed that the suggested measure can take both positive and negative values. Its positive value indicates that a region is resistant to crisis, whereas a negative one informs about the absence of resistance, i.e. sensitivity and vulnerability to crisis phenomena. 296 M. Markowska: A measure for regional… The growth rate of variables calculated against previous years (formulas (1) and (2) was used in the construction of the measure. It results from the assumption that the effect of crisis is manifested in a slowdown in growth or even a decline in the values of crucial economic factors. Destimulants are changed into stimulants in the course of growth rates calculation: 𝑤𝑖𝑗𝑡 =100(𝑥𝑖𝑗𝑡 𝑥𝑖𝑗,𝑡−1 −1) for stimulants, (1) 𝑤𝑖𝑗𝑡 =100(1− 𝑥𝑖𝑗𝑡 𝑥𝑖𝑗,𝑡−1) for destimulants. (2) At this point a conclusion can be drawn that in order to calculate an average rate a geometric mean rather than an arithmetic one should be used, however, w* values calculated below represent in fact the ratios of the rate and the median rather than the rate itself. The comparability of characteristics is obtained as a result of dividing the original rate values of variables changes (1) or (2) by the medians determined from spatio-temporal data modules (3). This transformation results in equal validity of the discussed characteristics. Such procedure maintains the original change rate sign and, moreover, the phenomenon of variables “flattening” by outlier values does not occur. Standardization (understood as achieving comparability) of changes of rates is performed by applying the following formula: 𝑤𝑖𝑗𝑡 ∗ = 𝑤𝑖𝑗𝑡 𝑀𝑒(|𝑤𝑖𝑗𝑡|) ⁄ (3) The measure of resistance to crisis is calculated as an arithmetic mean of the values of characteristics standardized by formula (3). The suggested measure takes the following form: 𝑅𝑅𝐶𝑖𝑡 =1 𝑚∑𝑤𝑖𝑗𝑡 ∗𝑚 𝑗=1 (4) where: i – object’s number (region), j – characteristic’s number, t – time unit number, m – number of characteristics, w* - standardized change rate, RRC – measure for regional resilience to crisis. The range of measure values does not have either upper or lower limit. It should be assumed that it corresponds to a rational opinion that, on the one hand, it is never so bad that it could not be worse and, on the other, it can always be better than it actually is. The measure of resistance can be transformed into the measure of sensitivity by multiplying it by (-1). STATISTICS IN TRANSITION new series, Summer 2015 297 4. Basic characteristics of RRC – preliminary assessment of results Economy, job market and households represent the areas of regional sensitivity to economic crisis. In order to perform the assessment of regional economic situations, in terms of their resilience or sensitivity to economic crisis, the following indicators were used in the study covering the period 2005-2011 (as of 31st October 2014 the information for 2012 regarding the data presented in values and necessary to calculate change rates was not provided by Eurostat database): – GDP in million PPS in a region (CR_GDP), – investments in million Euro in a region (CR_IN), – employment rate (as a percentage of professionally active population in 15-64 age group) (CR_ER), – unemployment rate (destimulant) (as a percentage of the total number of professionally active population) (CR_UR), – salaries in million Euro in a region (globally) (CR_S), – disposable income per capita in a household in PPS (CR_DI). The choice of variables was preceded by checking Eurostat database resources in terms of data availability, whereas the preliminary selection of variables was performed by assessing their changes, especially in 2009 against the previous years, among other things. The EU territorial units at NUTS 2 level constituted the base of regions covered by the assessment – the total of 264 regions (excluding Croatian and overseas Spanish and French regions – due to significant data gaps). In the dynamic assessment of changes the declines in 2009 against 2008 should be emphasized, since they were recorded in 250 regions (CR_GDP), 173 (CR_S), 212 (CR_IN), 202 (CR_DI), 205 (CR_ER). Moreover, for 171 EU NUTS 2 regions an increase in the unemployment rate was observed. It should also be emphasized that in the case of over 100 regions a decline in investments and GDP values was recorded also in 2008 (against the previous year). Simultaneously, further drops were observed in over 100 regions in the subsequent years for 159 and 110 regions with respect to employment rate and investments (127 and 120), along with an increase in the unemployment rate for 171 and 114 regions. In performing the assessment of regions in terms of RRC attention was also paid to the EU 15 regions (the so-called “old” EU) and the EU 12 regions – from the accessions in 2004 and 2007. The modules of medians of variables determined jointly in the entire period under analysis (dynamics in the period 2006-2011) were used in the standardization process and their values are presented in table 1. With reference to job market the attention should be paid to the median module of the unemployment rate which is several times higher than the employment rate. 298 M. Markowska: A measure for regional… The preliminary analysis of the obtained results indicates that 2009 represents the main crisis year in the EU NUTS 2 regions – the median is negative (median measure), as well as the mean value, and even the third quartile. In 2011 a group of weak regions was identified (Greek regions, in which a dramatic drop in salaries was recorded, among other things) which resulted in a strong left-sided asymmetry of the measure. The basic characteristics of the Measure for Regional Resilience to Economic Crisis are included in table 2. Table 1. Medians used in standardization Variable The median of change rate module in regions GDP change rate 4.54 Salaries change rate 4.15 Investments change rate 9.32 Household income change rate 3.22 Employment rate change 1.52 Unemployment rate change (destimulant) 12.50 Source: author’s estimations. Table 2. RRC characteristics Year 𝒙  Min 𝑸𝟎.𝟏𝟎 𝑸𝟎.𝟐𝟓 Me 𝑸𝟎.𝟕𝟓 𝑸𝟎.𝟗𝟎 Max SD As 2006 0.58 -1.55 -0.04 0.24 0.53 0.77 1.28 2.72 0.58 0.82 2007 1.02 -0.76 0.31 0.57 0.89 1.20 1.98 5.06 0.76 1.69 2008 0.48 -1.59 -0.90 -0.10 0.48 0.87 2.03 4.12 1.01 0.46 2009 -0.52 -3.12 -1.31 -0.80 -0.44 -0.15 0.16 1.23 0.60 -0.72 2010 0.36 -1.76 -0.54 -0.01 0.37 0.87 1.24 2.73 0.70 -0.17 2011 0.09 -5.34 -0.47 -0.11 0.27 0.73 0.97 3.29 1.26 -3.04 Source: author’s compilation. Picture 1 presents the distribution of regions in terms of RRC values. The same scale of horizontal axis allows one to “follow the moves” of the measure distribution. The highest diversification is observed for 2008, whereas the highest deviation is true for 2011 (the Greek group visible on the left side of the distribution). The effect of RRC distribution approximation by a normal distribution on one graph is illustrated on fig. 2. STATISTICS IN TRANSITION new series, Summer 2015 299 Figure 1. RRC distribution in the period Source: author’s compilation. Picture 3 presents RRC deciles (the first decile at the bottom and the ninth on the top). The line at the zero level stands for the division of sensitivity and resilience. Line 1 above represents the resistant regions. In the period 2006-2007 about 50% regions were included in this part. The 2009 crisis is well visible. Only slightly less than 10% of regions were placed on the positive side, thus only the best ones were resistant to crisis. R_06 R_07 R_08 R_09 R_10 R_11 -4 -3 -2 -1 0 1 2 3 4 0 20 40 60 80 100 120 140 160 180 200 Figure. 2. Approximation of RRC distribution by a normal distribution Source: author’s compilation. 2006 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 10 20 30 40 50 60 70 80 90 100 110 Number of regions 2007 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 20 40 60 80 100 120 Number of regions 2008 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 10 20 30 40 50 60 70 80 90 Number of regions 2009 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 20 40 60 80 100 120 Number of regions 2010 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 10 20 30 40 50 60 70 80 90 100 110 Number of regions 2011 -5,5 -4,5 -3,5 -2,5 -1,5 -0,5 0,5 1,5 2,5 3,5 4,5 5,5 0 10 20 30 40 50 60 70 80 90 100 Number of regions 300 M. Markowska: A measure for regional… Figure 3. RRC deciles Source: author’s compilation. On the basis of the analysis of numerical values of characteristics and the distributions of empirical values the division of RRC measure variability range into six classes can be proposed (see tab. 3). Table 3. The suggested RRC classes RRC value Class below -1 (-3) Strong sensitivity to crisis from -1 to -0.5 (-2) Average sensitivity to crisis from -0.5 to 0 (-1) Poor sensitivity to crisis from 0 to 0.5 (+1) Poor resilience to crisis from 0.5 to 1 (+2) Average resilience to crisis above 1 (+3) Strong resilience to crisis Source: author’s compilation. These classes allow for a more generalized assessment of sensitivity or resilience to crisis, as well as the quantification of these responses to crisis. 5. Results of the EU nuts 2 regions’ division into groups based on RRC values In each year of the study each region was assigned to one of the classes identified before. An even more general assessment than assigning to one of the six classes specified whether a particular region in a given year was resilient to crisis (a class coded with a plus), or sensitive to crisis (one of the classes coded with a STATISTICS IN TRANSITION new series, Summer 2015 307 BRIGUGLIO, L., KISANGA, E. J., (2004). Economic Vulnerability and Resilience of Small States. Islands and Small States Institute, University of Malta and Commonwealth Secretariat, Marlborough House, London. BRIGULIO, L., CORDINA, G., FARRUGIA, N., VELLA, S., (2006a). Conceptualising and Measuring Economic Resilience, https://secure.um.edu.mt/__data/assets/pdf_file/0013/44122/resilience_index. pdf [downloaded on May 28, 2014]. BRISTOW, G., (2010). Resilient regions: re-`place`ing regional competitiveness. Cambridge Journals of Regions, Economy and Society, 3, 153-167. CHAPPLE, K., BELZER, M., (2010). The resilient regional labour market: the US case. Cambridge Journals of Regions, Economy and Society, 3, 85–104. CHRISTOPHERSON, S., MICHIE J., TYLER, P., (2010). Regional resilience: theoretical and empirical perspectives, Cambridge Journals of Regions, Economy and Society, 3, 3–10. CLARK, J., HUANG, H.-I., WALSH, J., (2010). A typology of `Innovation Districts`: what it means for regional resilience, Cambridge Journals of Regions, Economy and Society, 3, 121–137. HASSINK, R., (2010). Regional resilience: a promising concept to explain differences in regional economic adaptability? Cambridge Journals of Regions, Economy and Society, 3, 45–58. MARKOWSKA, M., (2014). Ocena zależności między rozwojem inteligentnym a odpornością na kryzys ekonomiczny w wymiarze regionalnym – przegląd badań [Assessment of dependence between smart growth and resilience to economic crisis in regional dimension – research review, [in:] Gospodarka regionalna w teorii i praktyce [Regional economy in theory and practice], D. Strahl, A. Raszkowski and D. Głuszczuk (eds.), Research Studies of Wrocław University of Economics No. 333, Wrocław, 22–32. MARKOWSKA, M., (2015). The vulnerability of regions to economic crisis - measurement problems, [in:] Regional Economy and Policy. Territory and Cities, Hlavacek P., Olsova P. (eds.), Jan Evangelista Purkyne University in Usti nad Labem, Usti nad Labem, 104–112. MARKOWSKA, M., STRAHL, D., (2015). Inteligentny rozwój a podatność na kryzys ekonomiczny regionów Unii Europejskiej – próba oceny z wykorzystaniem modeli logitowych [Smart development and vulnerability to crises in UE regions – evaluation with logistic regression], Zeszyty Naukowe Uniwersytetu Ekonomicznego w Krakowie [Scientific Papers of the University of Economics in Cracow], Cracow (accepted for publication). 308 M. Markowska: A measure for regional… MASIK, G., RZYSKI, S., (2014). Resilience of Pomorskie region to economic crisis, Bulletin of Geography. Socio-economic Series, Volume 25, Issue 25, 129–141. MASIK, G., (2014). Uwarunkowania i instrumenty wspierające odporność gospodarczą miast i regionów, referat na konferencji na temat „Polityka miejska. Wyzwania, doświadczenia, inspiracje” (Warszawa 25-26.06.2013), http://www.euroreg.uw.edu.pl/media/prezentacje_konferencja_polityka_miejs ka/_2_grzegorz_masik.pdf [downloaded on May 28, 2014]. PERDAŁ, R., HAUKE, J., (2014). Czynniki rozwoju obszarów wzrostu i obszarów stagnacji gospodarczej w Polsce, [in:] Rozwój społeczno-gospodarczy a kształtowanie się obszarów wzrostu i obszarów stagnacji gospodarczej, P. Churski (ed.), Rozwój Regionalny i Polityka Regionalna, 25/2014, Instytut Geografii Społeczno-ekonomicznej i Gospodarki Przestrzennej, Bogucki Wydawnictwo Naukowe, Poznań, 69–88. PIKE, A., DAWLEY, S., TOMANEY, J., (2010). Resilience, adaptation and adaptability. Cambridge Journals of Regions, Economy and Society, 2010, 3, 59–70. REGIONS IN THE EUROPEAN UNION. NOMENCLATURE OF TERRITORIAL UNITS FOR STATISTICS NUTS 2010/EU-27, (2011). Series: Methodologies & Working Papers, European Commission, Luxemburg. ZAUCHA, J., CIOŁEK, D., BRODZICKI, T., GŁAZEK, E., (2014). Wrażliwość polskich regionów na wyzwania gospodarki globalnej, [in:] Gawlikowska- Hueckel K., Szlachta J. (eds.), Wrażliwość polskich regionów na wyzwania współczesnej gospodarki. Implikacje dla polityki rozwoju regionalnego. Oficyna Wolters Kluwer business, Warszawa, 206–244.