Now we consider life evaluations of respondents in Pakistan, covering the period 2013–2020 (Table 1). These values are taken from the reports of world value surveys.¹ From 2013 to 2018, Pakistan’s happiness score has risen by 26 points, making it the greatest gainer in the countries of the South Asian Association for Regional Cooperation (SAARC) and internationally one of the top 20 full gainers in 2020 (Helliwell et al., 2020). But it drops down to 4.934 and rank 105 among 149 happiest nations. There might be several reasons for this declining trend. There is a need of the hour to identify those factors that have dropped the happiness level in Pakistan. Therefore, this research aims to identify those factors that increase or decrease the level of SWB. The results of this study would assist policymakers in eliminating or controlling the variables responsible for people’s lower happiness levels.

Demographic determinants of SWB vary according to region, culture, and rank of countries in the previous studies. Age has a U-shaped relationship with happiness and LS in most of the studies (see Frey and Stutzer, 2000; Blanchflower and Oswald, 2004; Helliwell and Wang, 2011; Ngoo et al., 2015), and some studies found the non-linear relationship of age with SWB (Wunder et al., 2013; Proto and Rustichini, 2015). Relationship of family size, number of children, and marital status with SWB have also been considered vastly in previous literature as it affects teenage childbearing, educational attainment, and adult earnings. According to some previous studies, the number of children inversely affect happiness (Di Tella et al., 2003; Alesina et al., 2004; Ferrer-i-Carbonell, 2005; Gwozdz and Sousa-Poza, 2010; Ferreira et al., 2013, etc.), and that married people are happier than single people, who are happier than separated or divorced people (Clark and Oswald, 1994; Peiro, 2006; Dolan et al., 2008).

Much evidence is available on the significant and positive impact of income and education on SWB, which is also consistent across countries (e.g., Cuñado and de Gracia, 2012; Pereira and Coelho, 2013, etc.). Similarly, Dorn et al. (2007) also reported the positive impact of income on life happiness in 28 European countries. Alesina et al. (2004) found a positive linear relationship between income and happiness among Americans and European individuals. Results of developing countries are also homogeneous to developed countries. Meyer and Dunga (2014) reported a positive relationship between income and happiness and found that malnourished people are not happy in Africa. Ngoo et al. (2015) used the fifth wave of the Asian Barometer survey and found that the middle-income and high-income groups are happier with their living in comparison to the low-income group, and the results were more robust among the high-income group. Previous studies also suggest that financial wellness is also a major factor in improving wellbeing which is predicted by a high level of income, assets, own house, and low financial stress (Plagnol, 2011), which further increases the level of happiness and LS (Danish and Khan, 2021).

International variations in SWB are based on diverse living circumstances, particularly on the availability of social capital (Helliwell, 2003; Helliwell and Putnam, 2004). According to recent research, social capital is a significant driver of people’s happiness (Chang, 2009; Helliwell and Wang, 2011, etc.). Furthermore, another area of social capital that has gotten much attention is health and wellbeing (Putnam, 2000; Yip et al., 2007; Hoffmann et al., 2019; Danish and Khan, 2020). There is growing evidence of the positive impact of social capital on several aspects of people’s physical and psychological health (e.g., Lomas, 1998; Kawachi et al., 1999; Hawe and Shiell, 2000; Veenstra, 2000; Danish and Khan, 2020). Graham et al. (2011) suggested that having friends improves people’s happiness in Latin America. According to Vega (2016), spending more time with family and friends promotes happiness and improves people’s quality of life (QOL) in Mexico. In another study from Rwanda, Ngamaba (2017) found a positive relationship between trust level and network of friends with happiness. Al Bahar et al. (2021) found a direct association between the relationship with friends and family and happiness among older adults of Abu Dhabi, and Hoogerbrugge and Burger (2018) found a positive relationship between neighborhood social capital and life satisfaction.

There are also some variables on the country level that also affect people’s SWB of people like institutional quality, governance, and corruption. Previous literature suggests that political, economic, and judicial institutions are significantly related to the SWB of people at the country level. Trust in police increases nation satisfaction and happiness (Helliwell and Putnam, 2004). A study by Nizeyumukiza et al. (2020) demonstrates the positive and significant relationship between institutional trust and SWB in Indonesia. Youssef and Diab (2021) demonstrate that control of corruption and good governance raises the level of happiness in MENA countries. While on contrary, corruption and poor institutional quality reduce economic opportunity and raise inequality in society which further lead to lower SWB (Rothstein, 2010; Leon et al., 2013). Good governance, in general, decreases inequality and raises happiness (Ott, 2011; Kim and Kim, 2012; Danish and Nawaz, 2022). Crime may also affect the wellbeing of society as a whole and on an individual level. Crime reporting not only affects the victimized individuals but also increases society’s material and immaterial costs.

Determinants of SWB are widely discussed at the international level, but very few studies are available in Pakistan on SWB. For example, Shams (2014) investigated the socio-economic factors of SWB using the data of only 600 rural households from all provinces. Another study by Qaisar and Malik (2015) on public sector workers was conducted to determine their wellness and wellbeing in the context of Islamization. Hasan and Khan (2015) studied happiness as a determinant of the capability of being, functioning, and freedom in life. Jabeen and Khan (2016) empirically analyzed individual happiness in Pakistan by using data from world value surveys, while the effect of income on happiness in Pakistan is found in recent years. Recently, an attempt has been made to analyze the mediating role of health status between social capital and SWB by Danish and Khan (2020), the mediating role of financial satisfaction between income and SWB by Danish and Khan (2021), and the institutional quality and governance by Danish and Nawaz (2022) by using the same data in this research.

The world value survey (WVS) and the GALLUP survey collect data of 1,200 and 1,000 respondents on average from the whole country, which constitutes almost 15 respondents on average from each district, which is also unreliable for policy making at the district level. At the same time, this study will carry out after collecting a sample of more than 1,550 respondents from only four major districts of Punjab. Moreover, few studies use ordinary least square (OLS) regression on an ordinal scale of happiness and life satisfaction. Kim et al. (2015) analyzed the relationship between social capital and life satisfaction in Chinese and Korean elderly immigrants by using OLS, Hoogerbrugge and Burger (2018) also uses OLS while studying the neighborhood-based social capital on life satisfaction in Rotterdam, and Tsuruta et al. (2019) analyzed the relationship between social capital and happiness in Japanese communities and apply OLS with the ordinal outcome of happiness. All these mentioned studies did not apply the appropriate methodology in such kinds of analyses as the appropriate methodology on ordinal outcome dependent variable is order logit or order probit model (Cameron and Trivedi, 2005).

Moreover, no study is found at the national level which analyzes the determinants of SWB, including factors like corruption, governance, institutional quality, social trust, and health services, which play a vital role in improving happiness and life satisfaction among people of any country. In addition, this study also develops an index of three measures of SWB (happiness, LS, and life worthwhile) by using a weighted average method for the index. Thus, the present study fills the gap both at the national and international level in studying SWB, which will be carried through primary data from four major districts of Punjab, Pakistan (Lahore, Faisalabad, Rawalpindi, and Multan).

The present study is carried out by collecting the data of 1,566 households’ individuals in four major districts of Punjab, Pakistan (Lahore, Faisalabad, Rawalpindi, and Multan). The data are collected according to the population proportion of each district, randomly considering both rural and urban areas. More than 74% of respondents are men, while the remaining 25–26% are women. Respondents were asked to fill/tell the required information in the questionnaire. More than 1,000 respondents were interviewed from July 2018 to January 2019. While for the remaining 500 respondents, more than 900 questionnaires were distributed through the survey team with a response rate of 60%. The questionnaire was reviewed by three experts in this field, and data were collected after incorporating suggestions of expertise. The questionnaire includes information about household respondents’ personal and demographic characteristics: Economic factors of individuals and households like income, dwelling, debt, and so on, health and perceived qualities and satisfaction with institutions, perceived measures of life-influencing SWB like freedom, crime victim, and safety measures, and some variables of social capital, institutional quality, and perceived government effectiveness.

The following model is finalized to carry out this research.

where, i = 1, 2, 3,….., k “individuals.”

SWB_i* is the vector of subjective wellbeing measures [happiness, life satisfaction, life worthwhile, and an index of all three measures subjective wellbeing index (SWBI)]. Variables in Eq. 1 are defined in Table 2 along with sources.

According to previous research, SWB is defined as “a person’s cognitive judgment of life satisfaction, as well as their feelings and emotions, which economics and psychologists refer to as ‘happiness”’ (Helliwell and Putnam, 2004; Helliwell, 2006; Sarracino, 2013). But there is very little argument on measuring SWB through life worthwhile. Therefore, considering Dolan and Metcalfe (2012) and Deeming (2013), who advocated gaging each of the “Eudemonic” (Worth-While), “Evaluation” (Life Satisfaction), and “Evaluative” (Happiness) measures separately, this study also suggests gaging each of these measures. Following the WVS, respondents are rated on an ordinal scale from 0 (not at all happy/satisfied/worthwhile) to 10 (extremely happy/satisfied/worthwhile) for all aspects of SWB. Moreover, for individual measurement of SWB attributes, a SWBI is also constructed for everyone. Weight for happiness, LS, and worthwhile is assigned to the response of everyone, calculated from the sum values of each measure, and then this weight is multiplied by the actual response of an individual. Finally, all calculated happiness, LS, and worth values are added to construct the SWBI.

Finally, the following weights are assigned according to calculations estimated in Eq. 3 to get the SWBI.

The objective of this study is to analyze the determinants of SWB. As happiness, life satisfaction, and worthwhile of life have ordinal outcomes (0,1,2,…,10), ordered logistic regression is used to estimate all the models of wellbeing. The order logit model describes an unobserved latent variable (Cameron and Trivedi, 2005). The starting point is an index model with a single latent variable.

where “x” here does not include an intercept, with values of y_i = 0, 1, 2,…,10, being observed as yi* crosses gradually higher thresholds, which are also parameters to be estimated. An ordered logit model arises when u_i is logistic (or standard normal) distribution. Moreover, our analyses use the bootstrap standard errors with a maximum of 100 replications. It may be used to get cluster-robust standard errors in cases where clustering causes correlation within a cluster but has no effect on estimator consistency (Cameron and Trivedi, 2005).

The present research has also constructed the SWBI and considered for the analyses to analyze its determinants. SWBI is the discrete variable with limited values (0–10). Therefore, the tobit model is the suitable methodology for the model of SWBI (Cameron and Trivedi, 1986).

Table 3 shows the descriptive statistics for the sample study. The women are 26% of the sample, while the men are 74%. The average age of the respondent is 39 years, with an average education of 12 years and a family size of 6.46. Nearly 56% of sample respondents have an adjusted per-capita family income of less than Rs. 30,000 on average, while just 24% have an average per-capita household income of more than Rs. 45,000. Up to 43.5% of respondents said their health was good, while 5.7 and 1.9% said their health was poor or extremely poor. Only a small percentage of respondents were satisfied with the police, judiciary, and other government institutions. The range for “very high satisfaction” is 6–16%. Only about 16% of respondents believe that governmental institutions are free of corruption. Mean happiness, life satisfaction, and worthwhile were recoded as 6.59, 6.84, and 6.70, respectively, of the sample respondents.

Eighty-three percent of respondents completely trust family, while this ratio is only 24 and 33% for neighbors and friends, respectively. The ratio of trust in other people is very low; almost 78% of respondents do not trust people, which shows the lack of confidence in other people and a low level of social capital in Punjab. Only 7.8% of people are satisfied with the services of government hospitals, which indicates the poor performance or management in government hospitals. Therefore, the government must ensure the equity of health services in government hospitals.

Estimated outcomes of reported SWB among sample respondents (n = 1,566) are presented in Figures 1–3. About 8% of sample respondents reported a happiness level of only up to 20% (0–2), 33% reported from 40 to 60% (4–6), and 39% of sample respondents showed a happiness level between 70 and 90% (7–9), while 15.52% of sample respondents were completely satisfied with their lives.

According to Figure 1, most of the respondents were happy with respect to happiness level. Figure 2 shows that the life satisfaction of the respondents was observed high as compared with the average. So, the people of Pakistan are much happier than the average.

Figure 3 highlighted the facts about life worthwhile that show more percentage lie above the average proportion for life worthwhile.

This section presents the results of model 1, which is developed to estimate the determinants of SWB among households in Pakistan. Hierarchical order logit and tobit regressions are applied to analyze the effects of variables on SWB. Column 1 represents the variables used in the regression from Tables 4–8. In contrast, odd ratios for the model of happiness, LS, and worthwhile are reported from columns 2–4, respectively. The last column reports the coefficient values of the tobit model for the determinants of the SWBI. In the first step, only demographic variables are included in the analyses (model Ia). Variables regarding household finances are regressed over SWB in model Ib. Variables regarding the health status of individuals and institutional level variables are added in model Ic. In model Id, variables regarding some personal life experiences and perceptions regarding life are added. Variables of social capital are added in the model Ie. While in the last step, all the variables are added to the model as discussed in Eq. 1 (See Supplementary Appendix).

Results of demographic and personal characteristics are presented in Table 4, where age, gender, education, employment, marital status, and number of children are regressed over SWB. Contrary to the literature, age is negatively and significantly related to the happiness level. Older people above the age of 55 years reported the least level of happiness, while younger age people reported a high level of happiness. The odds of being happy at the upper age group level are less by 0.55 (OR = 0.45; p < 0.01) than the age group of people up to 25 years and by 0.17 than the age group of 26–35 years. Most of the aged people perceived poor health status which might be the reason for a low level of happiness. While, no significant relationship between age with LS and worthwhile and inclusive with overall wellbeing, as only the middle age group is significantly related to SWBI but not robust. Campbell et al. (1976) also predicted that demographic factors (age, gender, education, and marital status) only account for a 20% variation in SWB. The level of education is also significantly related to all measures of SWB. Individuals with higher education levels are happier, satisfied, and feel more worthfull about their lives compared to a very low level. There is no difference in the SWBI of university graduates as masters and above-level educated people have the same level of SWBI (β = 2.07; p < 0.01). Individuals holding certificates of accountancy or degrees in medical or engineering reported the highest level of SWB, and their happiness increased by odds of 1.16 (OR = 7.76; p < 0.01) to those holding other master-level degrees. Education improves the quality of thinking thus individual’s probability to get a job also increases which further raises the level of income which is directly proportional to happiness and LS (Ngoo et al., 2015).

Similarly, LS and worthwhileness of these people are higher by odds of 3.61 and 4.09 than the reference group (16-year education). Previous studies also find that higher education robustly affects the level of happiness and overall life satisfaction (Frey and Stutzer, 2000; Alesina et al., 2004; Jongbloed, 2018), while primary and secondary education have the least effect on SWB or sometimes have no effect (Davies and Hinks, 2010; Jongbloed, 2018; Nizeyumukiza et al., 2020). In Pakistan, it might be reasoning that highly educated people tend to get more employment or start their businesses to achieve a minimum standard of life to remain happy. Employment status also affects the SWB of people. Results of the present study also indicate that self-employed people are happier by odds of 0.59 than full-time employees, which are happier by odds of 0.48 than part-time employees. Overall, the wellbeing of self-employed people is high. At the same time, other groups (retired, unemployed/housewives) are not significantly related to SWB. The results of this study are also in line with previous studies and our hypothesis that self-employed and full-time employed people are happier with their lives (Headey and Wooden, 2004; Davies and Hinks, 2010), while unemployment negatively affects the happiness of people (Wang and Yang, 2016). Age, marital status, and the number of children are also statistically unrelated to SWB. However, up to three children, SWB decreases and then again increases. While married people are happier and more satisfied with their lives compared to single people, their overall wellbeing is also significantly higher than single. Marriage is generally seen as a strong predictor of wellbeing since it provides emotional, social, and economic support among married couples, hence increasing their happiness (Ndayambaje et al., 2020).

Moreover, women only show less satisfaction with their lives but are happier than men. However, this relationship is not significant at any level. Previous studies also show heterogeneous results on age, sex, and marital status. Few studies show the negative effect of age on happiness (Gwozdz and Sousa-Poza, 2010; Nikolaev and Rusakov, 2016). One of the major reasons for declining happiness levels with age is that health at an older age also tends to be poor in Pakistan, which causes stress and affects mental health.

Table 5 shows that holding agricultural land and a higher income level have a positive and significant impact on SWB. The happiness of the upper-income group (>60,000) is higher by 8.81 odds than the lowest income group (<15,000: reference category), and it is higher by 0.90 odds than the fourth quantile group (45,001–60,000). Similarly, LS of the upper-income group is higher by odds of 2.95, and worthwhile is higher by odds of 2.17 than the reference category (<15,000). Happiness and life satisfaction of agricultural landholders are also on the higher side by odds of 1.29 (OR = 1.29; p < 0.01) and 1.41 (OR = 1.41; p < 0.05), respectively. The results of this study are also consistent with hypotheses and many previous studies that find a positive impact of income and household assets on SWB (Alesina et al., 2004; Graham, 2011; Brown and Gray, 2016; Knoll and Pitlik, 2016; Krulichová, 2018, etc.). Improving the level of income eases the debt constraints and raises the consumption level by individuals or households which further raises the level of satisfaction and thus increases happiness (Pereira and Coelho, 2013).

On the other side, individuals with an outstanding loan to repay are less happy and satisfied with their lives by odds of 0.20 and 0.31, respectively, than those with no outstanding loan. Moreover, for borrowers who do not feel comfortable or are under financial stress, their SWB level further decreases. Previous studies also show that financial stress causes mental illness and lowers satisfaction with life and happiness (Brown and Gray, 2016; Aripin and Puteh, 2017, etc.). Overall, the SWBI of such people also decreased by 0.58 than the predicted values of SWBI (p < 0.05). Finally, people who live in rented houses have lower SWB than those who own their houses, but this relationship is significant only with life satisfaction (OR = 0.76; p < 0.05).

Health status, as expected, produces the same results as our hypothesis predicts. Poor health status is negatively related to the SWB (Table 6). The estimated SWBI of individuals with the worst level of self-reported health is at a minimum level (β = −3.42; p < 0.01). The reported happiness of individuals with poor health status is estimated to decrease by odds of 0.41 to those who reported good HS (OR = 0.13, P < 0.01). The level of LS and worthwhile is also badly affected by individuals’ poor SRH, and this relationship is significant at a 1% level. Poor health status is the major cause of low happiness, and life satisfaction as a person with poor health cannot enjoy the green of life. Moreover, individual with poor health restrict themselves socially which create stress and cause a lowering of level of happiness. The results of our study are also in line with hypotheses and previous studies in which good health is positively related to SWB while poor health is negatively related to SWB (Kahneman and Deaton, 2010; Garrett and James, 2013; Knoll and Pitlik, 2016). Hospitals’ satisfaction and services of hospitals also account for the wellbeing of people who avail or availed of the services when needed. A higher level of satisfaction with the services of hospitals (mostly visited by households) increases the wellbeing of individuals and vice versa. Similarly, people who were not satisfied overall with the situations of government or private hospitals reported a low level of happiness, LS, and worthwhile of life as well.

Moreover, respondents also show a low level of wellbeing if they face difficulties approaching hospitals, but this relationship is not statistically significant with the wellbeing (p > 0.1). However, interaction with several friends of barriers to the hospital improves the level of wellbeing, and relationship also becomes significant. In our society, people with strong social capital can easily approach doctors’ services with less cost minimizing the barriers to approaching health services, hence more wellness among people.

The institutional quality index (IQI) is also positively related to the SWB of people. Respondents who reported more satisfaction with police, court, and public institutions are happier and satisfied with their lives. SWBI is positively related to greater satisfaction with IQI (β = 0.41; p < 0.01). Similarly, the effective government index is also regressed over happiness and LS, which is also positively related to both factors of SWB, but a significant relationship is only found with LS. A one standard deviation increase in the satisfaction with government quality also increases the level of LS by odds of 1.10 (p < 0.05). Individuals who perceived better quality of institutions and government efficacy perceived more trust in such institutions and believe more support from institutions which raises the level of happiness and life worthwhile (Danish and Nawaz, 2022). Many empirical studies have found a strong positive relationship between trust in institutions and happiness. Dolan et al. (2008) suggested an important link between trust of association and wellbeing by conducting a literature review on SWB. Evidence from underdeveloped nations also suggests a positive correlation between institutional trust and SWB (Nizeyumukiza et al., 2020). Finally, respondents who perceived that the government was not involved in the corruption reported more levels of happiness as compared to those who perceived negatively about government corruption (OR = 1.48, p < 0.01). Tay et al. (2014) used data from a Gallup World Poll survey of 150 nations to find that perceived corruption lowers national GDP, lowering LS and pleasant feelings. Controlling corruption, on the other hand, raises national happiness (Li and An, 2020). Corruption has become a major problem that has affected the institutional trust in Pakistan as it also lowers the tax collection to help the poor and maintain the public health system (Habibov et al., 2019) which may reduce the level of SWB. While on the contrary, intuitional trust increases the willingness to pay more taxes by society, which not only increases the wealth of the nation but also increases government spending on the public which increases the level of happiness in the long term (Habibov et al., 2018).

Some variables other than the socio-economic factors may affect the psychological and mental state of happiness and life satisfaction. Some of these factors are also regressed over SWB, and the results are presented in Table 7. Results of the study depict that freedom is one variable that strongly affects an individual’s wellbeing level. More freedom of life choices led to increased happiness by odds of 1.49, LS by odds of 1.61, worthwhile by odds of 1.53, and all are significant at a 1% level (p < 0.01). In the past 2 years, crime victimization, either by respondent individual or family, also badly affect the life satisfaction of individuals. Individuals who also adopt some security measures in life, they always psychologically need to prevent risks. Hence, their wellbeing is also negatively affected as reported in our results (OR = 0.87; p < 0.01). Over the last decade, terrorism, civil war, and war against terrorism involving the country have also affected the state of people. People are mentally stressed with such situations in the country. Results of our study also confirm the hypothesis that for people who take the stress and worry about such situations in the country, their happiness and LS also decrease by odds of 0.06 and 0.09, respectively. Previous studies also find that self-victimization of crime or family significantly affects the level of happiness (Graham, 2011; Staubli et al., 2014; Krulichová, 2018, etc.).

Table 8 presents the results of social capital as a determinant of SWB. The number of friends, memberships with associations or alike, and strong trust in family and friends are all positively and significantly related to factors of SWB. A one standard deviation increase in the number of friends also increases the happiness level by odds of 1.34, LS by odds of 1.43, and worthwhile by odds of 1.27. It also increases the predicted probability of SWBI by 25%. Similarly, respondents registered with some voluntary organizations or associations more socially interact with other people and feel happier and satisfied with their lives. This relationship is also significant at a 1% level with all measures of SWB.

As discussed in the literature, trust is also an important determinant of SWB. People who trust their family, neighbors, and friends get the reward of trust in return, which connects them more strongly within and outside the groups. Such social engagement brings more happiness and, hence, more satisfaction in their lives. Moreover, the worth of such people also increases when they socially interact with others with the reciprocity of trust. Results of our study also depict that a decrease in the level of trust also decreases the level of wellbeing, but this relationship is significant only with the trust level of family and friends. There is growing evidence that social capital favors people’s physical and psychological health (Lomas, 1998; Kawachi et al., 1999; Veenstra, 2000; Ateca-Amestoy et al., 2014; Tsuruta et al., 2019; Yeo and Lee, 2019, etc.).