Детерминанты потребления домохозяйств в Азии: выявление региональных особенностей

Полный текст

Introduction Exceptionally rapid growth of consumption in Asian countries has been one of the most significant development trends in recent years (Kondratyev, 2020). Strong consumption, driven by rising incomes, stable inflation, and sustained remittance flows, underpins growth in Indonesia, Malaysia, Thailand, and the Philippines (Potapov, 2020). Actually, countries that are more successful in expanding their consumer potential (such as the Philippines, India, Indonesia, and, to some extent, China) exhibit relatively higher growth rates. The increase in household consumption is largely attributed to the rapidly expanding middle class, which forms the basis for effective demand. Rising incomes are accompanied by significant improvements in quality of life (Lee, Hong, 2012), and household expenditures are shifting from meeting basic needs to acquiring a more diversified basket of higher- quality and more technologically advanced goods and services (Hubacek, Guan, Barua, 2007). At the same time, the development of financial markets and digital technologies is making Asian markets more integrated, while the expansion of e-c ommerce, mobile platforms, and digital payment systems is stimulating consumer demand. The region’s medium- and long- term consumer potential is also shaped by the so- called low- base effect, a traditionally high household savings rate (Arapova, 2017; Kholopov, 2022), and relatively low-i ncome elasticity of consumption (Arapova, 2017). Despite the region’s enormous consumer potential, it remains significantly underperformed. While accounting for 47% of the world’s population, households in the ten selected Asian countries - China, Japan, India, Republic of Korea, Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam - accounted for only 17.4% of global private consumption in 2007. By 2023, this figure had increased to 24.2%, but it’s still much lower compared to the consumption in developed economies. In the EU and the US, which represent less than 9% of the world’s population, households account for over 44% of global consumption. In the context of declining external demand, the Asian region’s ability to boost consumption - and thereby maintain its role as a driver of global growth - depends on two factors: 1) the dynamics of key determinants influencing the growth of household consumption expenditure, and 2) the elasticity of consumer demand with respect to these determinants, considering the socio-e conomic and civilizational characteristics of Asian countries. The objective of this study is to quantify these elasticities for a group of the ten largest consumer markets in Asia and to compare the findings with global trends. This comparison aims to assess the region’s potential to emerge as a center of global consumption and, consequently, to conclude on the contribution of Asian consumption to regional and global economic growth. A substantial body of literature has examined disparities in consumption elasticity with respect to income across different country groups. It has been demonstrated that income elasticity of consumption is higher in both low- income and high- income countries, compared to middle- income countries (Diacona, Maha, 2015). However, with equal income growth, consumption increases more rapidly in developing countries than in developed ones (Mayer, 2013). After a certain income threshold is reached, the marginal sensitivity of consumption to additional income tends to decline (Zhao, Huang, Nie, 2022). Similar patterns may be also seen in Asia, although consumer behavior in the region is shaped by a much broader set of socio- economic factors. Despite growing academic interest in the Asian region, its shifting growth models, the exhaustion of export- led growth potential following the Global Financial Crisis, there remains a significant gap in comprehensive research on the determinants of household consumption and their elasticity. Most studies either focus on individual countries or are limited to the consumption of specific goods (e.g., food, fuel, electricity), and do not examine the regional characteristics of determinant impacts in comparison with global trends (Arapova, 2018; Rathnayaka, Selvanathan, Selvanathan, 2022; Guo, N’Diaye, 2010). This study aims to address this academic gap, comparing the elasticity of Asian consumption with global averages across key determinants. Methodological Approach and Data A regression model based on a panel dataset, tested using three approaches - ordinary least squares (OLS), fixed effects, and random effects models - seems to be the most appropriate method for achieving the stated research objectives. Within this study, a unified model of the determinants of private household consumption is tested for two samples of countries: (1) a broad set of countries representing different regions, and (2) a group of ten Asian countries, which constitute the focus of this research. The overall dataset comprises 4,719 observations (for a group of 143 countries over the period from 1991 to 2023), including 330 observations for the ten Asian countries over the same period. The selection of model indicators bas on key consumption theories and empirical research (Table 1). Table 1 Factors of the Regression Model and Their Theoretical Justification Indicator Name Model Variable Supporting Theories / Empirical Studies Expected Effect Household final consumption expenditure per capita (constant 2015 US dollars)1 HFCE - - GDP per capita (constant 2015 US dollars)1 GDP Absolute Income Hypothesis (Keynes, 1936) Positive Average GDP per capita over the previous 5 years (constant 2015 US dollars)1 GDP5 Permanent Income Hypothesis (Friedman, 1957); Empirical research by Campbell and Mankiw (Campbell, Mankiw, 1991) Positive Urban population (% of total population)1 Urban Empirical studies by Rosenthal and Strange (Rosenthal, Strange, 2004) Positive Population aged 15-64 (% of total population)1 Pop15_64 Life- Cycle Hypothesis (Modigliani, 1966; Modigliani, Ando, 1957) Positive Female population (% of total population)1 Female Empirical research by Morrison and Morrison (Morrison, Morrison, 2007); Malghan and Swaminathan (Malghan, Swaminathan, 2021) Mixed Inflation, consumer prices (%)1 Inflation (Leland, 1968); (Doepke, Schneider, 2006); (Lieb, Schuffels, 2022); (Weber, Hoang, D’Acunto, 2015) Mixed Services, value added (% of GDP)1 Services Empirical research by Francois and Hoekman (Francois, Hoekman, 2010) Positive Trade (% of GDP)1 Trade Empirical research by Francois and Hoekman (Francois, Hoekman, 2010); Porto (Porto, 2006) Mixed Financial Development Index2 FDI Empirical research by Bernanke and Gertler (Bernanke, Gertler, 1995); Bacchetta and Gerlach (Bacchetta, Gerlach, 1997) Positive Source: compiled by E.Y. Arapova from 1World Development Indicators (WDI), World Bank. https://databank. worldbank.org/source/world- development-indicators (accessed October 10, 2025) and 2Financial Development Index, International Monetary Fund https://db.nomics.world/IMF/FDI (accessed October 10, 2025). Based on the results of multicollinearity testing of the regressors, and due to the expected high correlation between current and permanent income levels, it was decided to test two separate model specifications, each incorporating the standard set of other influencing variables. ln HFCEit 0 1 lnGDPit1 2 lnUrbanit 3 ln Pop15_64it 4 ln Femaleit 5 ln(1 Inflationit1) 6 ln Servicesit 7 lnTradeit 8 ln(1 FDIit ) uit ; (1) ln HFCEit 0 1 lnGDP5it1 2 lnUrbanit 3 ln Pop15_64it 4 ln Femaleit 5 ln(1 Inflationit1) 6 ln Servicesit 7 lnTradeit 8 ln(1 FDIit ) uit. (2) Based on the results of the Wald, Hausman, and Breusch- Pagan tests for both specifications, the fixed effects model was identified as the most appropriate. Heteroskedasticity of the regression residuals was corrected using robust (heteroskedasticity- consistent) standard errors. The identified endogeneity problem, caused by reverse causality concerns the inflation variable (since not only does rising price levels influence household consumption behaviour, but also expanding consumer demand can provoke inflationary trends). This issue was addressed by lagging the inflation variable. Results and Discussion The econometric calculations demonstrate a high predictive power of the regression model: the model including the current income parameter explains 94.9% of the observations for the group of 143 countries and 98.9% for the Asian countries, while the model with permanent income explains 94.7 and 98.9% of the values, respectively (Table 2). Coefficients for most key factors in the regression model confirmed their statistical significance and the direction of influence consistent with established household consumption behaviour theories. The conducted calculations revealed a number of specific features of private household consumption in Asian countries compared to global trends. First, the population of the ten studied Asian countries reacts significantly less to changes in income levels compared to the global trend. While a 1% increase in current income globally leads to a 0.79% increase in per capita private consumption, for the group of 10 Asian countries this figure is only 0.66%. For permanent income, the figures are 0.69% for Asia versus 0.77% for the entire sample of countries. Notably, Asian consumption is relatively more responsive to growth in permanent income compared to current income. Asian consumers behave more cautiously and tend to plan consumption with consideration for long- term income expectations rather than one- time increases. Accordingly, assuming economic stability improves, the expansion of consumer demand will be more intensive, while the comparatively lower elasticity with respect to current income may restrain consumption surges in response to short- term income growth. With rising incomes, there is not only an expansion but also a diversification of the consumer basket. Dietary patterns change due to overall income growth and the expansion of the middle class. Consumers’ preferences shift toward higher- quality and luxury goods, and demand for innovative, creative products and content is increasing.Table 2. Table 2 Determinants of Household Consumption Expenditure at Global and Regional Levels Model and Dataset Current Income (143 countries) Current Income (10 Asian countries) Permanent Income (143 countries) Permanent Income (10 Asian countries) Number of obs 4.719 330 4.719 340 Number of groups 33 33 33 34 R-sq: Obs per group: within=0.9489 within=0.9888 within=0.9466 within=0.9888 between=0.9804 between=0.9954 between=0.9782 between=0.9947 overall=0.9494 overall=0.9893 overall=0.9473 overall=0.9892 LnGDP1/ LnGDP5 0.79139*** (0.01479) 0.659822*** (0.019064) 0.774181*** (0.015713) 0.685671*** (0.018165) LnPop_Urban 0.080463** (0.033682) 0.559581*** (0.02873) 0.0831** (0.036952) 0.492938*** (0.027538) LnPop15_64 0.209556*** (0.054691) -0.10319 (0.105888) 0.36124*** (0.045819) 0.012949 (0.093242) LnFemale 0.498245*** (0.079468) 4.06008*** (0.304756) 0.604555*** (0.091463) 3.575258*** (0.303996) Ln(1+Inflation1) -0.08885*** (0.02528) 0.554257*** (0.149592) -0.15281*** (0.026375) 0.454025** (0.184613) LnServices 0.407885*** (0.03003) 0.699735*** (0.048902) 0.387534*** (0.025985) 0.623928*** (0.046175) LnTrade -0.08978*** (0.0044) -0.01465* (0.008443) -0.07914*** (0.00493) -0.01942** (0.007982) Ln(1+FDI) 0.325003*** (0.063999) 0.172378** (0.085993) 0.409566*** (0.070377) 0.084912 (0.089281) _cons -3.15209*** (0.370323) -18.0605*** (1.564384) -4.01476*** (0.371642) -16.207*** (1.564668) Note. *, **, *** denote significance levels of 10%, 5%, and 1%, respectively. Source: E.Y. Arapova’s calculations based on data from sources listed in Table 1. Second, urbanization is a significant stimulating factor for consumption in Asian countries: amid its general significance, the coefficients for the group of Asian countries were 5 to 6 times higher than for the overall sample. The concentration of market participants in cities facilitates their interaction, benefiting both themselves and the local community as a whole, and causes shifts in the sectoral structure of the economy. A 1% increase in the share of the urban population in Asia leads to a 0.49 to 0.56% expansion in private household consumption. This means that urbanization can, in the medium and even long term, be one of the key drivers for achieving a consumption- driven growth model. In the Asian region, this is not merely an increase in the urban population share but a profound structural shift associated with the transition towards strong industrial and service economy. Urban growth, coupled with accelerated digitalization of urban infrastructure, ensures improvements in the quality and availability of a diversified range of goods and services, shaping new consumer habits. Considering current urban population growth rates, the stimulating effect on household consumption dynamics will be most evident in the short- and medium- term in China, Malaysia, Indonesia, and Thailand, and in the more deferred perspective - in India and Vietnam. Urbanization and technological development are driving the growing demand for smart technologies, eco- friendly products (shaping the trend toward sustainable consumption), and digital services. Third, price growth has the opposite effect on the dynamics of private consumption. While for the entire sample of 143 countries the inflation factor was significant but negative, for the dataset of 10 Asian countries a direct significant positive impact of inflation on household consumption was confirmed (both in the current and permanent income models). Price increases stimulate current consumption, whereas inflation expectations restrain deferred consumption. Experts explain this direct influence of price growth on consumption dynamics by the intertemporal effect, where, in anticipation of high inflation, current prices are perceived as lower than future ones, which stimulates current demand (Lieb, Schuffels, 2022). Expectations of rising prices may especially encourage spending on durable goods (Burke, Ozdagli, 2021). Price increases generate additional inflation expectations among Asian consumers, prompting them to make purchases “before further price increases”. Given the relatively lower financial security in some developing Asian countries and limited access to insurance and credit resources, inflation is perceived as a threat to real incomes, stimulating current expenditures. Moreover, this tendency becomes stronger when current income rises: with increasing prices, households are more willing to allocate a short- term income spike to additional consumption. Fourth, the negative effect of external trade openness was confirmed in all specifications of the regression model. For the Asian countries, this inverse relationship is primarily explained by the high export orientation of the largest economies in the sample (China, Malaysia, South Korea, etc.), since export orientation can stimulate economic growth at a certain stage but also increases savings rates, thereby restraining the expansion of private consumption (Athukorala, Suanin, 2025). Economies with inertial export orientation are more focused on capital accumulation to maintain competitiveness in external markets. Fifth, the development of the services sector plays an important role in stimulating consumption. In all specifications, this factor was positive and confirmed its direct influence in both the short and long term. Sixth, a key distinctive feature of consumer behaviour in this sample of 10 Asian countries is the significance of the financial development index only in the current income model. For the entire sample of countries, the financial factor ranks third in importance for stimulating private household consumption, contributing to its expansion by 0.32% and 0.41% in the current and permanent income models, respectively. However, this result should not be misleading. The set of the studied Asian sample includes regional economies that differ significantly in the level of financial development. This group can effectively be divided into three clusters: (1) Japan, South Korea, Malaysia, Singapore, and Thailand, for which the Financial Development Index ranges from 0.704 to 0.888; (2) China and India, with values of 0.634 and 0.534 respectively; and (3) the group of the least financially developed countries - Indonesia (0.364), the Philippines (0.379), and Vietnam (0.382), where the financial sector has not yet reached a level of maturity, capable of exerting a stimulating effect on private consumption. This imbalance explains the insignificance of the Financial Development Index in the permanent income model. Excluding these three countries, the coefficient’s significance is confirmed in both specifications, and its value is substantially higher than in the model for the 143 countries worldwide (Table 3). Table 3 Determinants of private household consumption in the current income model for 7 Asian countries (excluding Indonesia, the Philippines, and Vietnam) Model and Dataset Current Income Model Permanent Income Model Number of obs 231 231 Number of groups 33 33 R-sq: Obs per group: within=0.9924 within=0.9931 between=0.9898 between=0.9882 overall=0.9861 overall=0.9866 LnGDP1/ LnGDP5 0.563013*** (0.028628) 0.594105*** (0.02755) LnPop_Urban 0.440848*** (0.073252) 0.39753*** (0.07382) LnPop15_64 0.385403*** (0.119702) 0.470917*** (0.106241) LnFemale -0.35663 (0.569368) -0.54969 (0.540104) Ln(1+Inflation1) 0.911274* (0.536054) 1.007598* (0.526601) LnServices 1.170928*** (0.091516) 1.042475*** (0.099021) LnTrade -0.07449*** (0.012318) -0.07782*** (0.010891) Ln(1+FDI) 2.278968*** (0.180255) 2.138756*** (0.174699) _cons -4.20076* (2.262344) -3.28092 (2.206956) Note. *, **, *** denote significance levels of 10, 5, and 1%, respectively. Source: E.Y. Arapova’s calculations. Gender composition has a notable influence on the nature of private consumption. In all societies, women act as drivers of consumer demand; however, this factor manifests significantly stronger in Asia. The coefficient values for the sample of ten Asian countries in both models (current and permanent income) were 6 to 10 times higher. This trend is especially pronounced in the relatively less developed countries of the sample - Indonesia, Vietnam, and the Philippines - since the coefficients in samples including these countries are substantially higher than in the analysis limited to the seven more developed regional states, where this factor proved insignificant. Conclusion The largest Asian economies possess substantial but not fully performed consumer potential. The future development trends will depend less on the objective market capacity and more on the region’s ability to overcome institutional, infrastructural, and especially behavioral constraints, transforming them into drivers of regional demand. The traditionally high propensity to save among households remains a key barrier, which hinders the outpacing expansion of private consumption even while incomes rise. For consumers, long- term stability is more important than short- term income increases. Another limiting factor is the immaturity of financial infrastructure and the accumulation of financial issues in some countries of the region. Low financial accessibility, underdeveloped insurance, credit, and investment tools not only limit consumption but also strengthen the tendency to save as a way for households to protect themselves from economic instability. At the same time, the development of financial markets and overall institutional improvement of the financial environment can become key drivers for expanding consumer demand. If we attempt to rank the ten regional economies by their potential for expanding consumer demand in the future based on the dynamics of key determinants and identified elasticities, the following conclusions can be drawn: y Japan essentially lacks potential for further consumer market development; y In the short term (until 2030), the consumer market growth resources of Singapore and South Korea may be exhausted; y China will maintain its role as one of the key regional consumer markets by 2030 (mainly due to its large capacity), despite a negative forecast; y Malaysia, Thailand, and India will be centers of intensive consumer demand growth in 2025-2035; Indonesia, Vietnam, and the Philippines - in 2030-2040. Thus, the prospects for consumer demand growth in the region are delayed. Turning consumer demand into a driver of regional development requires simultaneous inflation control, financial infrastructure development, inclusion, and increased trust in financial institutions to reduce uncertainty and support sustainable long- term consumption. Public policy should aim to reduce gender inequality in access to resources and the labor market, support urban agglomerations, and work on reducing behavioral and institutional uncertainty to shift household economic behavior from saving to consumption.

Об авторах

Екатерина Яковлевна Арапова

Московский государственный институт международных отношений (Университет) Министерства иностранных дел Российской Федерации

Автор, ответственный за переписку.
Email: arapova_katrin@mail.ru
ORCID iD: 0000-0001-8765-9825
SPIN-код: 8700-6566

кандидат экономических наук, ведущий научный сотрудник, руководитель исследовательской программы «Мировая экономика. Цифровые финансы. Санкционная политика» Института международных исследований, доцент кафедры дипломатии, заместитель декана факультета международных отношений

Российская Федерация, 119454, Москва, проспект Вернадского, д. 76

Список литературы

Дополнительные файлы