1 Introduction

The well-being of its population is important for the sustainable development of any country. The quality of human capital defines many positive outcomes, including economic growth and development, promotion of democratic values, and a healthier and longer life. Despite similarities, Central Asian states have developed different economic and political trajectories. Oil-rich Kazakhstan is a relatively politically stableFootnote 1 and upper-middle income country, while Kyrgyzstan has often experienced political turmoil and is a lower-middle income country. In Kazakhstan, the GDP per capita has grown from $1,500 in 1992 to $ 9,139 in 2019.Footnote 2 The poverty rate has been reduced from 57% to 1996 to 8.6% in 2017, and the life span increased from 64 to 73 years.Footnote 3 Kyrgyzstan has also had similar trends albeit to a different degree. The GDP per capita has grown from $513 to $1,374, while the poverty rate dropped from 39.9 to 25.6% in 2006 and 2017 respectively. Life expectancy increased from 68.5 to 71.6 years by 2019.Footnote 4

Despite numerous studies, not much has been written on subjective well-being in Central Asia. Some research has been conducted on the impact of regional unemployment and inflation in Kazakhstan and other CIS states on subjective well-being (Kaluzhnova and Kambhampati 2008). Others showed that life satisfaction in Kazakhstan and Kyrgyzstan is higher than in the CIS states but lower than in Central and Eastern Europe. Scholars found that in addition to economic factors, societal cohesion and integration are important for well-being (Abbott et al. 2011). Abbott and Wallace demonstrated that such factors as being better off, having good health, education and control over one’s life are important contributors to life satisfaction. (Abbott and Wallace, 2012). Some ethnographic research on how people in Central Asia experience contentedness in their daily lives has been also done (Montgomery, 2013; 2015; Zanca, 2015; Roche, 2015; O’Neill Borbieva, 2013 2013; Werner et al., 2013). However, many of the studies have been conducted almost a decade ago and do not reflect the current situation in the region. None of the studies made a comparative analysis of the determinants of subjective well-being over time in Kazakhstan and Kyrgyzstan. Moreover, no research has been done on the predictors of change in subjective well-being (SWB) in the region. Although scholars looked at the determinants of well-being, they have not explained what factors predict the change in well-being. The study by Abbott and Wallace showed only the increase in well-being in Kazakhstan between 2001 and 2010. They, however, have not explained what factors predict that change. In addition, no study has focused on the change in well-being among the financially satisfied and dissatisfied citizens of two Central Asian states over the last decade.

This study fills the gap in the literature on change in subjective well-being in Central Asia. We can assume that there are contextual distinctions that might influence the SWB in both states. Using the data from (World Values surveys) WVS6 and WVS7, in this study we aim to explore if any change in well-being has occurred during the period from 2011 to 2018. We will identify what kind of change in well-being happened. In other words, have people become more satisfied or dissatisfied with their lives from 2011 to 2018? As our study suggests, well-being has changed in opposite directions in the two states. In Kazakhstan, well-being has declined from 2011 to 2018, despite the growth in GDP per capita, while in Kyrgyzstan it has increased. What factors predict this change (rise or decline) in subjective well-being? Finally, how have financially satisfied and dissatisfied citizens experienced the change of wellbeing in Kazakhstan and Kyrgyzstan over this period? We use the Blinder-Oaxaca decomposition method to determine the change in SWB. This study looks at both material and non-material factors such as income, freedom of choice, social capital, health, education, and other determinants existing in the scholarly literature to explain the change of SWB.

The paper is organized as follows. In the next section we focus on the literature review followed by the background section on Kazakhstan and Kyrgyzstan. Thereafter, we present the models and results of a regression analysis. In the discussion section we review our findings and explain the impact of determinants of declining SWB. The final section draws some conclusions.

2 Literature Review

The literature on the determinants of subjective well-being is very extensive. There is a general consensus among scholars that well-being is a primary purpose of human activity. According to Veenhoven (1991: 22), well-being is “a degree to which an individual judges the overall quality of his life-as-a-whole favorably,” (Frey & Stutzer, 2010, p. 12). Subjective well-being is an umbrella concept that includes different dimensions such as people’s emotional responses, judgments of life satisfaction and other domain satisfactions (Diener et al 1999). SWB can refer to different types of cognitive valuations including life satisfaction, the events which take place in their lives, their bodies, minds, etc. According to Dienere, “The term well-being is often used instead of subjective well-being because it avoids any suggestion that there is something arbitrary or unknowable about the concepts involved”. (Diener 2006). In this study, we refer to subjective well-being in terms of life satisfaction. “Life satisfaction represents a report of how a respondent evaluates or appraises his or her life taken as a whole. It is intended to represent a broad, reflective appraisal that a person makes of his or her life”. (Diener, 2006:401).

Scholars have explored various factors that might influence people’s well-being. One of the most examined determinants has been the level of income (Esterlin 1974, 1995; Deaton 2008; Tella et al. 2010; Guan et al. 2019; Bhuiyan and Szhulga, 2017). Most people throughout the world desire to achieve a high level of material wealth (Diener & Oishi, 2000: 185). For instance, Krugeman (1998:24) argued that pursuing economic prosperity can increase society’s well-being and brings happiness. To assess the influence of income level on life satisfaction, Veenhoven (1991) proposed a Theory of Livability based on the Maslow’s (1954) hierarchy of ‘needs’. According to the theory of ‘needs’, an increase in the level of income can explain the rise in subjective well-being as long as basic needs are met (Veenhoven, 1991). However, when the basic needs are met, an increase in income has only a limited positive impact on well-being (May, 2007).

Easterlin (1974: 113) proposed social comparisons as a reference point against which one’s income can be evaluated concluding that “relative status is an important ingredient of happiness”. Thus, subjective well-being is influenced by social comparisons, and people use relativistic judgements to asseslaters the level of their income in regard to the average income in their country. The Easterlin paradox shows that, despite a considerable rise in GDP per capita, the average happiness of a nation might not change over time, and a wealthier country is not necessarily a happier country. Some empirical studies revealed that at the individual level the reference income (comparison income) is negatively correlated with well-being (Sloane and Williams, 2000; Levy-Garboua and Montmarquette, 2004; Ferrer-i-Carbonell, 2005). The research on developed, developing and post-communist countries showed that well-being and income are positively associated in the short-term. However, in the long run, usually ten years or more, an increase in any country’s income does not improve the well-being of the population (Easterlin et al., 2010; Easterlin, 2001; Stutzer, 2004; van Praag & Ferrer-i-Carbonell 2004).

Financial satisfaction is also one of the factors that impacts SWB (Kahneman and Deaton, 2010; Ng and Diener, 2014). Ng and Diener (2014) define financial satisfaction as an individual’s self-evaluation of their financial situation (Ng & Diener, 2014). The scholars found that the association between financial satisfaction and SWB was stronger in richer than poorer nations. Another cross-national study also revealed a positive relationship between financial satisfaction and happiness (Hayes, 2014). Other scholars found that financial satisfaction is more important in poorer countries than in affluent societies (Delhey 2010). However, recent research has also revealed that financial satisfaction exerts a stronger effect on life evaluation in richer nations than in poorer ones.

Another cohort of scholars looked at the impact of non-material factors on well-being of the population. According to this strand of research, people focus on existential needs and prioritize economic and physical security under the condition of scarcity. However, after achieving certain material status that provides them with a sense of existential security, people “shift their emphasis from survival values toward self-expression values and free choice, which is a more direct way to maximize happiness and life satisfaction” (Foa, et al., 2008). Hence, free choice and control over one’s life is positively associated with happiness and valued as much as existential security (Johnson and Krueger, 2006; Inglehart & Welzel, 2005:140; Sen 2001). People in tolerant societies where they have relatively free choice, have higher levels of life satisfaction (Inglehart et al., 2008). Inglehart and colleagues argued that economic scarcity constrains people’s freedom of choice and sense of control over their lives, whereas living in developed countries contributes to a sense of greater freedom. People who live in wealthy and democratic societies are more satisfied with life because they also have more freedom over their lives (Inglehart et al., 2008; Johnson & Krueger, 2006). This implies that in developed societies, post-materialist needs matter more to life satisfaction than material values (Delhey, 2010; Inglehart, 1997). People highly value trust in their socio-political environment including trust in colleagues, government incumbents, public services, and neighbors. Social involvement and engagement with others, including active participation in the public decision-making, delivering public services and politics, make people more satisfied with their lives (Helliwell, 2006; Frey & Stutzer 2010).

Scholars propose that in low-income countries other non-economic factors such as religiosity, in-group solidarity and national pride are closely associated with happiness. Helliwell and Putnam (2004), drawing on empirical evidence from 3 different samples, revealed that social capital is strongly and positively associated with subjective well-being. Their research has confirmed other findings. Indeed, frequent and intense social interactionism is positively correlated with life satisfaction, while loneliness has a negative impact on self-reported happiness (Stack & Eshleman, 1998; Frey & Stutzer, 2010; Stavrova et al. 2012; Cracolici et al. 2014; Kudebayeva et al. 2022).

Other socio-demographic variables such as age, health, education and gender have also been explored. Scholars have found a U-shaped relationship between age and well-being. The level of life satisfaction is the lowest in one’s middle age (Gerdtham and Johannesson, 2001; Blanchflower and Oswald 2008; Steptoe et al. 2015; Bhuiyan and Szhulga 2017; Abbott and Wallace 2012). Good health is positively correlated with well-being (Abbott and Wallace 2012). According to research, women demonstrate a higher level of self-reported happiness than men. Women are more emotional and tend to have lower aspirations in comparison with men, which makes them feel happier. However, Frey & Stutzer (2010: 66) propose that the difference in happiness among men and women is not significant and tends to disappear in later decades.

Inglehart and his associates (2008), using a sample of 88 countries between 1981 and 2007, revealed that almost all post-Soviet societies reported a lower level of subjective well-being compared to other countries such as India, Bangladesh, Nigeria, Mali, and Ethiopia (Inglehart et al., 2008; Guriev & Zhuravskaya, 2009). Inglehart and his associates (2008; 2013) interpreted the sharp decrease in subjective well-being as a consequence of the loss in national pride following the collapse of political, economic, belief systems and weakness of religion that led to the spiritual vacuum in the society. Other researchers have explained the decline in well-being as a result of the poor public goods provision (Guryev and Zhuravskaya, 2009; Bartolini et al. 2013).

Another important factor, which has been explored by scholars is political freedom. The evidence suggest that individuals, who live in mature and young democracies with political freedoms, are happier than those who live in countries with restricted political freedoms (Berggen et al. 2018; Brjonskov and Rode 2020; Bjronsnkov and Tsai, 2015; Altman et al. 2017; Diaz-Serrano and Rodríguez‐Pose 2012; Haller and Hadler 2006; Helliwell and Huang 2008; Veenhovetn 1995; Weitz-Shapiro & Winters 2011). For instance, Frey and Stutzer (2000) found that in those cantons where people actively engaged in politics and local government they were happier than the regions with low level of participation. In another study, scholars (2005) found positive correlation between political participation rights and levels of life satisfaction. (Frey and Stutzer 2005). Another studies also showed that greater political freedoms and opportunities to take part in political processes lead to higher levels of life satisfaction (Owen et al. 2008; Tavita 2008). Bjornskov et al. (2010) found that political factors are important in middle and high-income countries but not much in low-middle income countries suggesting that democratization and political freedoms are good for SWB when a country reached a certain level of economic development. Other scholars looked at the countries in transition finding that political rights and liberties are positively linked with life satisfaction (Hayo 2007).

The study of CIS countries between 1995 and 2014 showed that the economic factor and social capital were powerful explanatory variables of happiness (Dimitrova, 2017). In Kyrgyzstan, divorced, unemployed, and older people had a low level of self-reported happiness compared to other social groups (Namazie and Sanfey, 2001). In Russia, unemployment had a very big impact on the level of well-being in the early 1990s (Selezneva, 2011). The determinants of well-being in Kazakhstan have also been examined. Kaluzhnova and Kambhampati found that the size of one’s dwelling and Kazakh ethnicity are important predictors of well-being. Other factors tested in the study included regional unemployment rate and inflation. It was found that in 2006 unemployment had no impact on well-being of unemployed people, while those employed experienced some satisfaction (Kaluzhnova and Kamhampmati, 2008). Another study on Kazakhstan, Kyrgyzstan, Georgia and Armenia showed that in 2001 the level of satisfaction with life was comparable with that of in other CIS states but lower than in the states of Central and Eastern Europe. Scholars aimed to evaluate the impact of political and economic changes that countries were going through.(Abbott et al. 2011). They found that the population in Kyrgyzstan had higher levels of happiness than other states. They explained this by the level of trust towards relatives and friends, higher levels of control over life, and cultural factors. In addition, it was shown that not only economic factors but also social integration (measured in terms of general trust) and societal cohesion are important for life satisfaction in these countries (Abbott et al. 2011). Replicating the same study in 2010 in Kazakhstan, Abbott and Wallace found that the level of life satisfaction over nine years has increased, such factors as being better off, good health, education and control over one’s life are important predictors for well-being. However, life satisfaction is not determined by age and gender (Abbott and Wallace, 2012). Most recent research showed the impact of social capital and people’s satisfaction with public services on SWB in Central Asia and Kazakhstan (Kudebayeva et al., 2022; Serikbayeva and Abdulla 2022). However, none of this work has shown how well-being changed over time and what factors contributed to this change in well-being in two states and across different groups within countries.

3 Background

3.1 Kazakhstan

Kazakhstan is one of the most dynamic political and economic countries in the Central Asian region. It is rich in terms of oil and gas resources. During the last decade from 2010 to 2020, Kazakhstan has experienced various socio-economic and political challenges including the decline of economic growth and the devaluation of the tenge - national currency - in 2015 by 30%. During this period, Kazakhstan faced a fall in the oil price, on which Kazakhstan’s economy depends heavily on. The over-reliance on the export of oil, which is almost 60% of GDP, makes Kazakhstan extremely vulnerable to external shocks, which in turn affect citizens’ well-being. According to WVS, in 2018 56,3% of people were satisfied with their life, while 11.8% were dissatisfied. Other respondents, 31,9% were neither satisfied, not dissatisfied.Footnote 5

In the 1990’s Kazakhstan experienced a deep economic crisis and radical cuts in public spending (Cook, 2011). Compared to the 1990’s, the living standards of the population in Kazakhstan have improved significantly in the 2000s. The poverty rate has been reduced to 2.75% in 2015, from 5.5% to 2011; however, it has slightly increased in 2019 up to 4,2% by national poverty line (i.e. subsistence minimum). The poverty line for Kazakhstan was updated by World Bank from the previous $5.5 in 2011 PPP to a new $6.85 level based on 2017 PPP. The poverty rate has been reduced to 14.29% in 2018 from 19.17% to 2011 and increased in the midst of the pandemic till 25.6% and expected to decline to 15.5% in 2022 according World Bank. According to official statistics, the unemployment rate fell to 4.6%Footnote 6 in 2019.Footnote 7 In turn, life expectancy has increased from 64 years in 1996 to 73 years in 2018.Footnote 8 Notably, all these numbers are produced before the Covid pandemic. The situation after COVID-19 has changed. According to the survey conducted by the Friedrich Ebert Foundation (FES) in 2018, 69% of respondents emphasized the size of salaries as an important factor in their life. This demonstrates that material values are pivotal for the citizens of Kazakhstan. Indeed, better-off citizens are more satisfied in their lives and feel happier than those who do not have an adequate financial status.

Politically, Kazakhstan is a presidential republic that has a two-chamber parliament. Kazakhstani society can be characterized as rather passive and apathetic. The percentage of people who are members of different associations and organizations is exceptionally low and varies from 0.8 to 4.4%.Footnote 9A small percentage of the population is interested in politics − 41.8%.Footnote 10

3.2 Kyrgyzstan

Kyrgyzstan is one of the most democratic but politically unstable countries in the Central Asian region. Economically, Kyrgyzstan belongs to the lower-middle income countries. It has experienced popular riots in 2005, 2010 and 2015 years due to the fraudulent elections and interethnic conflicts. As such, compared to Kazakhstan, Kyrgyzstan can be characterized as having a more vibrant political life and more active civil society. However, Kyrgyzstan is one of the poorest countries in the region. Its economic growth has been changing from − 0.5 to 11% in 2010 and 2013 respectively. The fluctuations in economic growth have been tightly connected to a rapidly changing political situation. In contrast to Kazakhstan, Kyrgyzstan held several elections of presidents (2005; 2009; 2010; 2011; 2017; 2021) that were accompanied by popular uprisings. During this turmoil period, the country also experienced interethnic clashes between Kyrgyz and Uzbeks. In 2015, the newly elected president, S. Japarov introduced constitutional changes which increased his presidential powers. At the same time, the country has one of the most vibrant civil societies in the region. Kyrgyzstan has score 3 out of 7 in terms of civil societyFootnote 11 In 2017, around 23,859 NGOs were registered in the republic.Footnote 12

All political changes have been accompanied by ups and downs in economic growth and a high level of political instability. After the Tulip revolution in 2005, economic growth decreased from 7% to -0.2%, followed by another decrease from 8.4% to 2008 to -0.5% in 2010.Footnote 13 In addition, it is estimated that 25–41% of Kyrgyzstan’s GDP relies on remittances which may also depend on the economic situation in such countries as Russia and Kazakhstan (WDI, 2020).Footnote 14Nevertheless, despite this volatility, Kyrgyzstan has experienced positive economic growth starting from the 1990’s.

We observe that the poverty rate has decreased from 36.8% to 2011 to 22.4% in 2018 by national poverty line. It shows that every fifth person lives below the poverty line in Kyrgyzstan. The COVID-19 pandemic has worsened the situation, increasing the percentage of the poor up to 25.3% in 2020 and to 33.3% in 2021. By the World Bank poverty line $3.65 in 2017 PPP terms and poverty rate has been declined from 19.6% to 2011 to 13.22% in 2018 and increased to 18.7% in 2020. Life expectancy also increased in the country over the last two decades from 68.5 to 1991 to 71.4 in 2018.

4 Data and Methodology

To determine the factors that influence well-being in Kazakhstan and Kyrgyzstan, we extracted data from the World Values Survey (WVS) 6 and 7 conducted in 2011 and 2018 for the two Central Asian countries: Kazakhstan, Kyrgyzstan. The sample sizes for each country consist of 1,500 households in 2011 and 1,276 in 2018 for Kazakhstan and 1,200 for KyrgyzstanFootnote 15. The percentage of missing data is less than 10% for Kyrgyzstan and Kazakhstan;Footnote 16 this does not raise concerns of the serious biases of the estimates (Allison, 2001; Little and Rubin, 2002). Descriptive statistics are presented in Appendixes A1, A2, and A3.

This WVS dataset provides a wide range of questions on different issues including a proxy of Subjective Wellbeing (SWB). In our study the dependent variable is life satisfaction rather than happiness. The rationale behind the choice of this dimension is the following. First, perceptions of life satisfaction are stated on a ten-point scale, while happiness is measured only by a four-point scale. The former therefore offers better and more segregated evidence than the latter. In addition, it is usually held that happiness delivers a more emotional degree of well-being. In contrast, life satisfaction replicates a more rational assessment of well-being and is therefore seen as more reliable (Diener, 2006).

In this study, we measure life satisfaction as “a subjective appreciation of one’s life-as-a-whole”, or global satisfaction, rather than as a number of domain specific measures in which people state their degree of satisfaction with different aspects of their lives (Veenhoven, 2007; 2012). A common measurement of subjective well-being is the following question in WVS: “All things considered, how satisfied are you with your life-as-a-whole now?” The respondents must choose on a scale from 1 to 10 with 1 “dissatisfied” and 10 as “satisfied”.

In our model we have included such important explanatory variables of life satisfaction as freedom of choice, financial satisfaction, and household income. As previous research suggests, all these factors are crucial for people’s well-being. The variable freedom of choice is measured on a scale from 1 (“none at all”) to 10 (“a great deal”) to indicate how much freedom of choice a respondent feels over his/her own life. The variable financial satisfaction is constructed based on the following question “How satisfied are you with the financial situation of your household?” The answers to this question range from 1 to 10, where one stands for complete dissatisfaction and 10 stands for the complete satisfaction. The household income variable is measured by an ordered scale of income intervals.

We also have employed proxies to measure social capital. Three variables have been used as proxies for social capital: index of civicness, trust in people, and social participation. The index of civic cooperation is based on answers to questions: if the following are acceptable: “claiming government benefits which you are not entitled to”, “avoiding a fare on public transport”, “cheating on taxes if you have the chance”, or “accepting a bribe”. Answers to these questions range from 1 (always justifiable) to 10 (never justifiable), so that the larger values stand for stronger norms of civic cooperation. We have applied factor analysis to construct a composite index of civicness.

In addition, in our model we have included the level of trust in people as one of the dimensions of social capital. It is considered that trust in people positively influences people’s well-being. This variable is based on answers to the following question: “Generally speaking, would you say that most people can be trusted, or that you can’t be too careful in dealing with people?” The binary variable is set to 1 if the respondent answers positively, and 0 otherwise.

Another variable in our model is social participation used also as proxy of social capital. It is based on the following questions: “Could you tell me whether you are member of ‘X’ type of organization?” The answers to this question range from 0 (don’t belong), 1 (inactive member), 2 (active member). The list of organizations includes religious organization, sport or recreational organization, art, music or educational organizations, labour union, and other organizations. We measure associational activity with a dichotomous variable set to 1 if a respondent is actively engaged or performs unpaid voluntary work for at least one of the mentioned groups or associations, and give 0 otherwise.

In our model, we included control variables for the degree of non-tolerance toward neighborhood. The dummy variable is set 1 for the ‘yes’ answer, and 0 otherwise. Factor analysis has been utilized to construct a composite index of non-tolerance to the neighborhood. High values on the index imply a low level of tolerance. Finally, we also included a set of social-demographic controls such as age, gender, marital and employment status, number of children, reported health, ethnicity and the size of city.

In this study, our goal is to explore and explain the trend of SWB in Central Asia by quantifying the comparative significance of the changes in the predictors of life satisfaction. For this purpose, we have used the Oaxaca-Blinder decomposition. This approach permits us to decompose the well-being gap between the wave 6 and wave 7 observations and to detect the extent to which changes over time in the levels and in the coefficients of life satisfaction predictors explain the well-being gap. The Oaxaca - Blinder decomposition was introduced by Blinder (1973) and Oaxaca (1973) to study discrimination between men and women in the labor market. The decomposition approach has been applied in other fields, including research on subjective well-being (Helliwell and Barrington-Leigh, 2010; Sarracino, 2010).

The decomposition approach permits us to evaluate group variances in an outcome variable by dividing its differential in two parts; the explained one, accounting for differences in observed determinants of the population and the unexplained one, measuring the variances in the coefficients between groups. In this study, the decomposition permits us to identify how much of the overall change in the average subjective well-being between two waves can be attributed to changes in the set of characteristics (the explained part) and to changes in how these characteristics are estimated (unexplained part).

A drawback of the Oaxaca-Blinder decomposition, however, is that the unexplained part captures also the potential effects of differences in any unobserved variables. The decomposition can be presented as follows:

$$\varDelta SWB={\left[E{X}_{W6}-E({X}_{W7}\right]}^{{\prime }}\bullet {\beta }^{*}+\left[{E\left({X}_{W6}\right)}^{{\prime }}\bullet {\beta }_{W6}-{\beta }^{*}+{E{X}_{W7}}^{{\prime }}\bullet ({\beta }^{*}-{\beta }_{W7})\right]$$
(1)

explained part unexplained part.

where SWB is the difference in average subjective well-being between the wave 6 (W6) and wave 7 (W7) of observations, E(x) is the wave average of a vector of explanatory variables measured at the wave 7 and wave 6 of the period of observation, βW6 and βW7 are vectors of coefficients, and β* is a vector of non-discriminatory coefficients to compute how much each group variables explains the overall difference of means. This decomposition shows how much of the difference in mean welfare between the two waves can be attributed to differences in the waves’ characteristics and how much to differences in the returns to these characteristics. The vector of explanatory variables are: gender, age and age squared, marital and employment status, income, financial dissatisfaction, subjective health, freedom of choice, trust in others, index of civic cooperation, participation in groups and associations, and size of the city.

5 Results

5.1 Kazakhstan

We perform the decomposition of Eq. (1) and thus determine the magnitude of the mean subjective wellbeing differentials between Wave 6 and Wave 7 and investigate how much of these differentials is due to differences in characteristics and how much is due to differences in the returns to these characteristics.

What we observe is the difference in subjective well-being between the two waves as the mean for Wave 6 minus the mean for Wave 7. Hence, a positive difference implies that subjective well-being is higher in 2011 compared to 2018, and a negative difference implies the opposite. The Oaxaca-Blinder decomposition illustrates a statistically significant reduction in subjective well-being for the period between 2011 and 2018 in Kazakhstan (see Table 1). In other words, well-being in Kazakhstan has declined significantly over the eight-year period. The endowments effect captures the differences attributed to such various characteristics on subjective well-being. In turn, the coefficient effect captures the differences attributed to the outcomes given the same characteristics.

The third component shows the interaction between differences in prevalence and differences in coefficients. The difference in the subjective-well-being between Wave 6 and Wave 7 is mostly due to the coefficients effect. If the two waves had the same returns to their characteristics, the differential in the subjective well-being value would be 121.3% of the current difference of 0.239. The disaggregated results reveal that the largest contributors to the subjective well-being differential are the returns to financial satisfaction, state of health and income of household as well as the constant. Other significant contributors to the subjective well-being differential are social participation, non-tolerance to the neighborhood, freedom of choice, number of children, widowed, retired and students (see Table 1). As the model shows, financial satisfaction returns, non-tolerance to the neighborhood, freedom of choice and being widowed have a negative impact on the gap in well-being, which implies that, had Wave 6 and Wave 7 the same characteristics and the same returns to their characteristics (including the same constant), the subjective well-being differential would be negative, which means higher subjective well-being in Wave 7. Notably, after income and financial satisfaction, the next most important contributor to the decline of subjective well-being in 2018 is the decrease in freedom of choice. As evidence suggests, people perceive that they have less choice in their lives in 2018 than they had in 2011 and this, in turn, negatively affects their well-being.

Table 1 Oaxaca-Blinder decomposition between wave 6 and wave 7 for the whole sample, Kazakhstan

Since financial satisfaction plays a crucial role in the reduction of subjective well-being between 2011 (wave 7) and 2018 (wave 7), we decomposed subjective well-being between two periods among those who are financially satisfied and dissatisfied (see Tables 2 and 3). As the table below shows, subjective well-being has significantly declined from 7.547 to 2011 to 5.741 in 2018 among the financially dissatisfied, which is about 24% (see Table 2). Notably, 68.9% of this decline is explained by the coefficients effect and 48% by endowments.

Table 2 Oaxaca -Blinder decomposition between wave 6 and wave 7 for financially dissatisfied portion of population
Table 3 Oaxaca -Blinder decomposition between wave 6 and wave 7 for the financially satisfied population Kazakhstan

At the same time, the Oaxaca-Blinder decomposition reveals that SWB has significantly increased by 10.2% among financially satisfied people (see Table 3). The main contributors to this increase are the endowments effect, which is 54.6% of total increase 0.72. In addition, the coefficient effect also contributes significantly, comprising 28.1% of the total increase. The detailed decomposition indicates that the main contributors to the difference in SWB between 2011 and 2018 for the financially satisfied people are: having freedom of choice, a state of health, trust in people, household income, number of children, being widowed, having a university degree, and being retired.

5.2 Kyrgyzstan

The results of the decomposition of the life satisfaction gap in Kyrgyzstan during 2011–2019 are presented in Table 4. In 2011 the average level of life satisfaction was 6.96 on a 10 points scale, and in 2019 it increased to 8.36. Overall, the subjective well-being in Kyrgyzstan increased by 20.12% during the 2011–2019 period. Around 20% of the variation is explained by the changes in endowments between the two waves. In other words, if the levels of each explanatory variable had not changed over time, the well-being differential would have been 20% smaller. The remaining 80% of the life satisfaction gap remains unexplained, that is to say it is related to variations in preferences between periods. The main predictors of this increase in subjective wellbeing are non-tolerance to the neighborhood, household income, financial satisfaction, state of health, marital status, and type of employment. The non-tolerance from 2011 to 2019 declined by 9.12%, while financial satisfaction increased by 17.18%, number of divorced individuals increased by 157.7%, and full and part-time employment also increased (see Appendix A1).

Table 4 Oaxaca-Blinder decomposition between wave 6 and wave 7 for the whole sample in Kyrgyzstan

The results of the decomposition of the life satisfaction gap among financially dissatisfied people in Kyrgyzstan during 2011–2019 are presented in Table 5. Subjective well-being increased by 16.51% among financially dissatisfied people. Both endowments and coefficients (change in preferences) statistically significantly influence the gap of subjective wellbeing between two periods. Mainly, the endowments of the financially dissatisfied people reduced the gap in SWB, meaning an negative effect on the increase of SWB. The main factors that affected the change in the gap of subjective well-being among financially dissatisfied are: trust in others and index of civicness; also, the freedom of choice. The decline in trust in others and an increase in index of civicness from 2011 to 2019 reduces the gap, meaning negatively affecting the increase of SWB. An increase in freedom of choice and almost the same values of non-tolerance to the neighborhood between two periods through returns to endowments (unexplained part of decomposition) increases the gap of SWB during 2011–2018, meaning a positive influence in the increase of SWB. Being self-employed and a housewife through the change of preferences also increases the gap of SWB, meaning a positive impact on the increase of SWB.

Table 5 Oaxaca -Blinder decomposition between wave 6 and wave 7 for financially dissatisfied portion of population in Kyrgyzstan

The results of the decomposition of the life satisfaction gap among financially dissatisfied people in Kyrgyzstan during 2011–2019 are presented in Table 6. The SWB increased among financially satisfied by almost 20%. Some 24% of this change is through the endowments of the individuals and remaining 76% percent is the unexplained part of the decomposition. The decline in trust, the decline in household income, and the reduction in index of civicness reduces the gap in subjective wellbeing, meaning a negative effect on an increase of SWB between these two periods. The decline in nontolerance to the neighborhood, increase in freedom of choice, the improvement in state of health, and an increase in number of children increases the gap in SWB, having a positive impact on increase of SWB. Other factors, such as age, marital status and type of employment are significant factors on the increase in SWB.

Table 6 Oaxaca -Blinder decomposition between wave 6 and wave 7 for the financially satisfied population in Kyrgyzstan

6 Discussion

Our models therefore show very interesting and important results across the two states. There were different trends in Kazakhstan and Kyrgyzstan despite the fact that these countries have many things in common. The two states have similar cultures, traditions; people speak common Turkic language. In terms of political system, the two states have developed similar institutions, albeit Kyrgyzstan is considered more democratic due to its active civil society. At the same time, we observe the decline in well-being in Kazakhstan and increase in life satisfaction in Kyrgyzstan from 2011 to 2018. Also, the difference is across various social groups. In Kazakhstan despite the overall decline in well-being, financially satisfied people experienced an increase in SWB, while financially dissatisfied people had the opposite trend. In Kyrgyzstan both groups experienced an improvement in well-being. These differences are predicted by the different political and economic dynamics in two states over this period.

In both countries, as the models demonstrate, important factors that predicted the change in SWB are freedom of choice and financial satisfaction. In Kazakhstan, freedom of choice (through endowments) increased the gap in SBW between 2011 and 2018, thus negatively influencing SWB. People have experienced a decrease in freedom of choice and this has negatively predicted their well-being. In contrast, freedom of choice in Kyrgyzstan has increased the gap of SWB between 2011 and 2018 thereby contributing to better well-being of the population. In other words, those people who feel that they have a choice and control over their lives are more satisfied. This is an interesting finding since previous research has mostly focused on freedom of choice in democratically affluent societies rather than in developing or poor states (Inglehart et al., 2008). Our study has shown that freedom of choice can be an important predictor of well-being in authoritarian states as well.

Another factor, financial satisfaction, has positively predicted the difference in subjective well-being between 2011 and 2018. Those people who could improve their financial status feel more satisfied than those who did not. Previous research has also shown that financial satisfaction might lead to better well-being since people feel more secure than those who are not financially satisfied.

In Kazakhstan, as the model suggests, social participation and a university education through returns to endowments have a positive effect on the gap in SWB. The more people are educated, engaged in various organizations, and the more they interact with other people, the more they are satisfied with their lives. This finding corroborates previous research on the importance of social capital. Although, in general, people in Kazakhstan have a low level of membership in various civil society associations, nevertheless engaging in professional organizations and interaction with other people positively predict people’s satisfaction with life.

In Kyrgyzstan, other factors such as one’s state of health, non-tolerance to neighborhood, and part-time employment through returns to endowments contributed to the gap in well-being. All those factors increased the gap in SWB thereby improving well-being of the people. People who work only part-time as well as those who have become more tolerant toward their neighbor and with better health have experienced an improvement in their well-being from 2011 to 2018. Health has been an important predictor of people’s satisfaction with life.

We also consider how other factors predicted the change in SWB through coefficients. We found that in Kazakhstan such factors as household income and health also have a positive impact on the difference in SWB. Since Kazakhstan experienced an economic crisis and the devaluation of tenge, for many citizens income can make a difference. It is known that people in Kazakhstan spend a large share of their income on food. On average, the monthly salary of Kazakhstani citizen is 162,673 KZT in 2018Footnote 17. A rise in income level can significantly improve people’s satisfaction with life. Those people who perceive that they are in good health and have adequate income are more satisfied with their life. In contrast, in Kyrgyzstan, household income decreased from 2011 to 2018 and thus negatively influenced the well-being gap. For Kyrgyz people whose income has been reduced for that period, the satisfaction with life has been also reduced.

In Kazakhstan, good health can significantly improve people’s well-being while poor health might deteriorate life conditions. The number of children has a positive impact on the gap of SWB, which implies that an increase in the number of children negatively predicts SWB. This finding is in line with previous research in other contexts. A large number of children might decrease the level of life satisfaction due to the raised level of responsibilities and lack of time on the part of parents. According to the FES survey, 95% of citizens of Kazakhstan said that family is the most important value. Furthermore, many people in Kazakhstan believe that family should be large, 62.3% at the 18–29 age old and 71% of people at the age 51–60 years old.Footnote 18 On average, there were three children in a family in Kazakhstan in 2019. However, a large number of children can be related to more stress and anxiety and thus lead to a lower level of SWB. The results of decomposition of SWB between the Waves demonstrated that such variables as people at the retirement age and students also positively predicts the change of life satisfaction through returns to endowments. This implies that SWB has decreased for students and people who are retirement age, controlling for other factors.

At this time, in Kazakhstan such factors as freedom of choice, non-tolerance, being widowed, and financial satisfaction negatively predict (through coefficients) the gap in SWB. It means that these factors positively impact the increase of SWB, since they reduce the change in preferences over the eight-year period. Among these factors, financial satisfaction and freedom of choice have the largest impact on the gap in SWB than all other variables. Both factors decrease the difference in SBW and thus positively predict life satisfaction. People who are financially satisfied and have more freedom of choice can experience higher levels of satisfaction with life.

In Kyrgyzstan, financial satisfaction increased over time and thus positively predicted the change of well-being of Kyrgyz people. Other factors that positively predicted the difference in SWB include being divorced and having a part-time job. The non-tolerance toward neighborhood declined over time and contributed positively to life satisfaction in Kyrgyzstan.

However, the most interesting results of our model include the difference between the financially satisfied and financially dissatisfied individuals who experienced different levels of SWB. In Kazakhstan for the financially satisfied people, SWB has increased over time, while for the financial dissatisfied individuals, SWB has decreased. In Kyrgyzstan, however, SWB increased for both groups. This is an interesting difference between the two states. The dissimilarity between the two states can be explained by distinct political and economic situation. In Kyrgyzstan, despite the volatility, we observed economic growth and an increase in political freedoms. During this period more people were involved in political activities, including the elections of presidents. It could contribute to a higher level of satisfaction in life. In Kazakhstan, the same period has been characterized by a similar political situation and the worsening in people’s rights and liberties although the country has experienced the moderate economic growth rates compared to the period of boom 2000–2008. In the case of Kazakhstan, the SWB increased for financially satisfied people due to endowments, through better characteristics, such as education, health, marital status, employment, and income. However, for Kyrgyzstan SWB increased for those who were financially satisfied due to change in preferences (or returns to endowments).

7 Conclusion

As this study shows, subjective well-being has declined in Kazakhstan and increased in Kyrgyzstan from 2011 to 2018. This is an important difference between the two Central Asian states. Our study has shown that there are some important factors that influenced the change in SWB. As it was shown, freedom of choice and financial satisfaction have been important in both states and contributed to the SWB. During this period, people experienced a decline in freedom of choice and decrease in terms of control over their lives due to the economic crisis, devaluation of tenge and political instability. This has negatively affected their satisfaction with life. Our study contributes to previous research by showing that freedom of choice can also determine the well-being of people in authoritarian states. This factor might be powerful not only in democratically affluent societies but also in countries with authoritarian or semi-authoritarian regimes. To improve their well-being, people should have more freedom of choice in their lives. Financial satisfaction also positively predicted SWB in both states. It shows that this factor can predict to what extent people are satisfied with their lives. Indeed, the research in other contexts also shows the importance of education for well-being, while the lack thereof leads to a lower level of well-being and poor health.

One of the important findings of this study is the difference in terms of life satisfaction across financially satisfied and dissatisfied groups. We have observed that from 2011 to 2018, SWB decreased for the financially dissatisfied and increased for the financially satisfied population in Kazakhstan. This disaggregated analysis shows that the decline in SWB is not a universal trend for the entire population of Kazakhstan. Rather, it might vary from one group to another. A uniform approach applied by other scholars glosses over important nuances that might exist on the ground. Hence, it is important to disentangle various factors to have a better understanding of change in SWB in various contexts. This finding also confirms previous research that financially satisfied citizens experience higher levels of life satisfaction. In Kyrgyzstan, we found that both groups have experienced better well-being. It could be explained by the relative economic growth in the country during this period and more political rights and freedoms for people.

Thus, the governments of Central Asia should pay more attention to better education and provide more freedoms to people in order to improve their well-being. At the same time, people also should engage more in public life since higher levels of social participation can improve their well-being. In addition, citizens of Central Asia should be aware of their level of tolerance towards other people. As such, the government and people should aim to construct a liberal and inclusive society to improve their lives.