A Comparative Study of Saving Behaviour Between India and China

Abstract

This study attempts to understand the determinants of saving behaviour using the Global Findex micro-database of India and China. Further, this study has also tried to identify the gender gap in saving behaviour for both the countries. Empirical (pooled logistic regression) results suggest that being rich, educated, employed and old favour saving than others. Women are more prone to save informally than men. The main contribution of this article is the analytical comparison between India and China, which demonstrates that in terms of saving Chinese adults are ahead of Indian adults. However, informal saving is more prevalent in India. The gender gap in saving behaviour is higher in China than in India. Our research also discovered that China’s age saving pattern is U-shaped, that is, younger and older are more likely to save than the middle-aged, which contradicts the standard life cycle model whereas this model holds for India.

Keywords

Saving behaviour age gender India China

I. Introduction

Saving is a vital link between today’s decision and tomorrow’s quality of life. It provides individuals with opportunities to transfer resources from the present to the future, so that, they can overcome unforeseen emergencies and also secure their future needs. Thus, personal savings ensure the lifetime economic stability of individuals and their households and it also allows people to improve their standard of living through wealth accumulation (Gokhale, 2000). However, individuals do not follow a consistent pattern of saving throughout their life. All people, on the other hand, do not save in the same way. The life cycle hypothesis model argues the same and identifies a hump-shaped saving pattern (Modigliani, 1986). It expounded that children or elderly people save less as their productivity is low, whereas the saving rate is higher among the middle age (working-age) population due to their high level of productivity. Subsequent studies have further expanded their research and they found that, apart from age, other socio-economic and demographic factors, such as income, occupation, household size, education, dependent children, marital status, ethnic background, etc. have a significant influence on the absolute level of household savings and savings rate (Hua & Erreygers, 2019; Lugauer et al., 2019; Nwosu et al., 2019; Schultz, 2005). These factors are the driving forces that shape the level of savings and savings rate of an individual or a household. As a result, different countries’ saving habits differ (Demirgüç-Kunt et al., 2018; Tobing, 2012). In this light, the purpose of this study is to examine whether the socio-economic factors influence the decision to save, the decision regarding different modes of saving and the decision regarding different motives behind saving. Further, a comparative analysis will be done in this regard between Asian giants; China and India, which are two outstanding comparable countries in the world. So, we are actually trying to examine the impact of various socio-economic factors on the different aspects of the saving behaviour of the individuals at the micro-level in these two countries. China and India are not only the two most populous countries, but also have large economies. The huge population provides a huge labour force in these countries, which act as assets in these two countries. They have quickly risen to prominence as significant economic forces. China and India are very important members of BRIC countries and have experienced a series of economic reforms since 1978 and 1991, respectively (Ang, 2009). But, when it comes to GDP, China outperforms India (World Bank, 2021). Even in terms of savings rate, India lags far behind China (World Bank, 2021).

The household savings rate in China was not so high from the beginning. Ang (2009), in his paper, mentioned that both India and China had similar household savings rates in 1953. After that, over the next two decades, the savings rate increased dramatically in India but not so much in China. The sudden spurt in the savings rate of China has been observed since 1978 due to their far-reaching economic reforms. In 1984, China overtook India and since then, has maintained a significantly higher savings rate. The introduction of the one-child policy is another great strategy adopted by China in the 1970s, which resulted in a rapid reduction in fertility rates and less young-age dependency (Ang, 2009). Later, several studies have found the low age dependency ratio as the predominant driver of a higher savings rate in China (Hung & Qian, 2010; Lugauer et al., 2019). Although India’s age dependency ratio is decreasing over time, it has a smaller impact on household savings in India than it does in China (Ang, 2009). However, the higher savings rate does not necessarily mean the higher number of people is saving. Earlier studies have mostly focused on determining the factors that influence the household savings rate in India and China. A higher savings rate is beneficial for economic growth (Domar, 1946; Harrod, 1939; Lean & Song, 2009; Solow, 1956; Zhang et al., 2016), but all adults must have the opportunity to save in order to improve their standard of living. Evidence suggests that the percentage of adults that save is also substantially higher in China than in India (Demirgüç-Kunt et al., 2018). Hence, here in-depth analysis is required to understand which individual characteristics drive saving, do they affect both country’s adults in the same way.

People can save in a variety of ways, i.e., either formally or informally or both. Formal saving means saving at a formal financial institution, whereas informal saving refers to saving using a savings club or a person outside the family etc. Formal saving necessitates the opening of a formal account, however, having a formal account does not always indicate formal saving. Evidence suggests that in comparison to India, China appears to have more formal account holders who save formally (Demirgüç-Kunt et al., 2018). Furthermore, people go about saving money for various motives, such as for old age, education, business, etc. It’s not reasonable to expect everyone to save for the same reasons and in the same mode in every country. The urgency to save for a particular motive varies from country to country.

Given this background, our research aims to evaluate the determinants of saving behaviour and make a comparative analysis between India and China. Our study differs from the earlier studies in some respects. Firstly, most of the studies examined the factors influencing the absolute level of household savings and savings rate; in this article, we solely have focused on individual-level data to determine the socio-economic factors that influence the decision of saving for an individual. Secondly, apart from saving, we also have identified determinants of modes of saving and various motives behind saving. Thirdly, concerning saving determinants, earlier studies have compared China and India at the macroeconomic level, in our study these comparisons have been done at the microeconomic level.

Gender inequality is a global issue and its elimination has recently become a fundamental aspect of the Sustainable Development Goals 2030. In this regard, we have extended our research to examine the gender gap in terms of different aspects of saving behaviour, such as the decision to save, the decision on different modes of saving and the decision on different motives behind saving, in both the countries. This article is unique in this sense that no earlier studies had compared India and China in terms of the gender gap in saving behaviour. These results will enable us to tell whether the Indian females are lagging behind the Indian males more than Chinese females from Chinese males.

The rest of the present article is set out as below. Section II furnishes the literature survey. Section III is dedicated to the data and methodology. Section IV represents empirical results and a discussion of our main estimation. Section V concludes with policy implications.

II. Literature Survey

An extensive review of earlier literature is essential, for new areas so far unexplored may be identified and studied in depth. However, a concerted attempt has been made to analyze the earlier research on micro and macro determinants of saving and gender differences in saving.

Clancy et al. (2001) examined the effects of financial education on saving motives for the poor in Individual Development Accounts (IDA) for the USA. IDAs are saving programs that provide an institutional structure for saving and offer significant benefits, especially to low-income people. They observed that general financial education has substantial effects on savings. ARIÇ (2015) focused predominantly on macroeconomic determinants of saving for Middle East countries and the empirical results suggest that income, money supply and government expenditures impact savings negatively, whereas young population and inflation are positively associated with saving. De Vos et al. (2020), in their study, shed light on understanding the state of household savings in South Africa’s low-income groups. They discovered that low-income households in South Africa genuinely save less; however, the government’s grant to households has a favourable influence on saving levels. Nwosu et al. (2019), make a significant contribution towards addressing the influence of socio-economic characteristics on the level of household savings for Nigeria. They demonstrated that land ownership, living in good sanitary conditions (they considered it as an indication of high income), single-person households, education above secondary, female-headed households, age of the head and rural dwelling have a positive and significant effect on savings. While, household size, age of the head above 73, poor electricity supply (it was a proxy for poor infrastructural development), employment in agriculture, polygamy, widowhood and separated households affect household savings negatively. Hua and Erreygers (2019) exclusively emphasized household characteristics to assess their effect on household saving rates in Vietnam and found income, the ethnic background of the household head, the educational level of the household head as key influencing factors of saving. A recent study, by Akram (2021) for Pakistan, used micro-data to identify the effects of a range of socio-economic variables on household savings. Findings suggest that income, living in a nuclear family, house ownership, receiving remittances from abroad and being involved in agriculture have a positive implication on household savings, whereas the dependency ratio exerts a negative effect on saving.

Several empirical studies have been carried out to understand the factors influencing household saving, particularly in China and India. A study by Horioka and Wan (2007) is one of them, examined the macroeconomic determinants of household saving rates in China. They found two factors, such as lagged saving rate and income growth rate, have a positive significant impact on the saving rate. In many cases, they observed that the real interest rate exerts greater influence on the saving rate, and in some cases, impact of the inflation rate has been identified. An interesting comparative study between India and China on the determinants of household saving has been done by Ang (2009). The estimated results for both countries support the view of the life cycle hypothesis, that is, higher income growth encourages household saving, whereas the reverse is true for higher age dependency. Here also, inflation rate is observed to be an important factor that promotes household saving. The most important finding of this study is that Chinese households behave differently as compared with Indian households. In the long run, an increase in expected pension benefits promotes household saving in India, but in China, opposite is observed. The argument regarding determinants of saving was put forward by Samantaraya and Patra (2014), for India and their results revealed that age dependency has a favourable effect on household saving, while high inflation has a negative impact, which is very much contradictory to the results reported in Ang’s (2009) paper. Furthermore, the positive influence of GDP and the negative impact of real interest rates on saving have been demonstrated in this study. The consensus that emerges from some recent empirical studies (Chamon & Prasad, 2010; Lugauer et al., 2019) is that in China the saving pattern is U-shaped rather than inverted U, that is, younger and older relatively save more as compared with middle-aged. This finding contradicts the traditional view of the life cycle model regarding the relationship between age and saving. Furthermore, Lugauer et al. (2019) also revealed that dependent children are negatively associated with household saving rates, that is, Chinese families with lower dependent children have a higher rate of saving. Bohini et al. (2021) established a new dimension through an empirical investigation for India, that is, the millennial youth working-age group has a beneficial influence on gross domestic saving in the long run, but, an adverse impact in the short run.

Gender is a well-known factor that substantially affects saving behaviour. It has been discovered in numerous studies that the saving behaviour of males and females is not homogeneous. Seguino and Floro (2003) conducted their research for a set of semi-industrialized countries and observed that women’s propensity to save is higher than men’s. Similar evidence is also manifested in Lee and Pocock’s (2007) work for South Korea. Ghosh and Hom Chaudhury (2019) investigated the presence of a gender gap in formal saving exclusively for India. They explored the fact that women are 4.9% less likely to save formally as compared with their male counterparts and this gender disparity is completely driven by socio-economic differences between men and women.

III. Data and Methodology

Data Used

In order to carry out our research, we have used the World Bank’s 2017 Global Findex micro-database for India and China. This database provides individual data on more than 200 indicators of financial inclusion, such as account ownership, saving and borrowing. All indicators are derived from survey results. This survey has been carried out by Gallup, Inc. in association with its annual Gallup World Poll. Global Findex data cover almost 150,000 people in more than 140 economies worldwide, representing over 97% of the world’s population. Using randomly selected, nationally representative samples, 3,627 people in China¹and 3,000 people in India have been questioned in the survey. The nature of variables has been widely discussed in section ‘Variable Used’ and Table A1.

Variable Used

The dependent variables used in our analysis, such as saving, formal saving, informal saving, saving for old age and saving for farm or business purposes. As all the responses are in binary form, hence, we have created a dummy variable for each dependent variable (see Table A1). The explanatory variables are country, gender, age, education, income and employment. We developed four dummy variables for income, that is, income level 1 (poorest 20%), income level 2 (second 20%), income level 3 (third 20%) and income level 4 (fourth 20%). The fifth richest quintile is the base dummy. We have used two dummy variables for education: secondary education and tertiary education. The base category is primary education or less. Age and age square (to consider the non-linear relationship with dependent variable) has been introduced. We also have created dummy variables for gender and country. Detailing of explanatory variables has been given in Table A1.

Empirical Model

This article uses the following Equation (1) to examine how individuals’ characteristics are associated with saving, modes of saving and saving motives, and also to make a comparative analysis between India and China. We have performed Pooled logistic regression to estimate the following equation.

\begin{matrix} S_{i} = α + β C o u n t r y_{i} + γ_{1} G e n d e r_{i} + σ_{1} A g e_{i} + σ_{2} A g e_{i}^{2} + \\ δ_{1} I n c o m e l e v e l 1_{i} + δ_{2} I n c o m e l e v e l 2_{i} + δ_{3} I n c o m e l e v e l 3_{i} + \\ δ_{4} I n c o m e l e v e l 4_{i} + θ_{1} T e r t i a r y e d u c a t i o n_{i} + θ_{2} S e c o n d a r y e d u c a t i o n_{i} + \\ ϕ_{1} E m p l o y m e n t_{i} + γ_{2} C o u n t r y * G e n d e r_{i} + σ_{3} C o u n t r y * A g e_{i} + \\ σ_{4} C o u n t r y * A g e_{i}^{2} + δ_{5} C o u n t r y * I n c o m e l e v e l 1_{i} + δ_{6} C o u n t r y * \\ I n c o m e l e v e l 2_{i} + δ_{7} C o u n t r y * I n c o m e l e v e l 3_{i} + δ_{8} C o u n t r y * \\ I n c o m e l e v e l 4_{i} + θ_{3} C o u n t r y * T e r t i a r y e d u c a t i o n_{i} + θ_{4} C o u n t r y * \\ S e c o n d a r y e d u c a t i o n_{i} + ϕ_{2} C o u n t r y * E m p l o y m e n t_{i} + ε_{i} \end{matrix}

(1)

where S_i indicates saving, modes of saving (formal and informal²) and saving motives (saving for old age and saving for farm or business purposes) and i is the index for individuals. Apart from the country, gender, age, education, income and employment, we also have incorporated interaction dummies, such as Country*Gender, Country*Age, Country*Age², Country*Secondary education, Country*Tertiary education, Country*Incomelevel1, Country*Incomelevel2, Country*Incomelevel3, Country*Incomelevel4 and Country*Employment in this equation to make a comparative analysis between India and China. Here, the Wald statistic has been used to understand the significance of models (Wooldridge, 2012). To obtain a more prominent picture of saving behaviour, we have constructed a cross-tabulation on saving and socio-economic variables for both India and China separately (Tables A2 and A3).

We also have adopted ‘The Blinder–Oaxaca decomposition technique’ (Sinning et al., 2008), including bootstrap to understand the significance of the gender gap in saving behaviour for both India and China. We have computed the gender gap in saving, modes of saving and various saving motives by estimating the following generalized decomposition equation (pooled model):

\begin{matrix} {\overset{─}{Y}}_{M} - {\overset{─}{Y}}_{F} = {E_{β^{*}} (Y_{i M} | X_{i M}) - E_{β^{*}} (Y_{i F} | X_{i F})} + \\ {E_{β_{M}} (Y_{i M} | X_{i M}) - E_{β^{*}} (Y_{i M} | X_{i M})} + \\ {E_{β^{*}} (Y_{i F} | X_{i F}) - E_{β_{F}} (Y_{i F} | X_{i F})} \end{matrix}

where, M = male and F = female. ${\overset{─}{Y}}_{M} - {\overset{─}{Y}}_{F}$ denotes the mean outcome difference between males and females.

IV. Results and Discussion

Preliminary Analysis

We started by looking at the descriptive statistics of all the variables used to proceed with our analysis. Table 1 presents the mean and standard deviation for samples of India and China.

Table 1.
Summary Statistics

Variable China
India

Observation Mean SD Observation Mean SD

Saving 3,608 0.45 0.50 3,000 0.34 0.47

Modes of saving

Formal saving 1,626 0.64 0.48 1,005 0.57 0.50

Informal saving 1,626 0.07 0.26 1,005 0.28 0.45

Various saving motives

Savings for farm/business purposes 1,626 0.26 0.44 1,005 0.28 0.45

Savings for old age 1,626 0.48 0.50 1,005 0.34 0.47

Variable	China	India
Saving	3,608	0.45	0.50	3,000	0.34	0.47
Modes of saving
Formal saving	1,626	0.64	0.48	1,005	0.57	0.50
Informal saving	1,626	0.07	0.26	1,005	0.28	0.45
Various saving motives
Savings for farm/business purposes	1,626	0.26	0.44	1,005	0.28	0.45
Savings for old age	1,626	0.48	0.50	1,005	0.34	0.47

Source: Calculated from World Bank’s Global Findex database (micro-data), 2017.

Note: As all variables are qualitative by nature, here, mean denotes the proportion of positive response.

Evidence reveals that more adults save money in China than in India. Forty-five per cent of Chinese adults reported saving money, while 34% of Indian adults did. The majority of savers opted to save formally in both the countries. Informal saving is less in China, however, its significance in India cannot be understated. Statistics show that 28% of Indian savers save money informally. The main motivation for saving in both China and India is ‘for old age’.

Socio-economic Factors Influencing the Saving, Modes of Saving: Broad Comparison Between India and China

Table 2 reports the results of pooled logistic regression analysis for determining factors affecting saving, modes of saving, that is, formal savings, informal savings and also to make a comparative analysis on saving determinants in China and India.

Table 2.

Determinants of Saving, Modes of Saving

	Pooled Logistic Regression
	Saving	Formal Saving	Informal Saving
Country	3.2543*	2.9080*	−1.2052*
Gender	0.0608	0.0128	0.4869*
Age	0.0639*	0.0435*	0.0899*
Age square	−0.0005*	−0.0003	−0.0010*
Secondary education	0.4756*	0.5122*	0.1105
Tertiary education	0.9633*	1.2429*	−0.4386
Income level 4	−0.2707**	−0.3185**	−0.6697*
Income level 3	−0.3967*	−0.4595*	−0.9456*
Income level 2	−0.7052*	−0.6964*	−1.3158*
Income level 1	−1.0946*	−1.2182*	−1.1318*
Employment	0.6105*	0.5376*	0.6365*
Country*Gender	−0.3435*	−0.3271**	−0.6561*
Country*Age	−0.0992*	−0.0725*	—
Country*Age square	0.0008*	0.0005***	—
Country*Secondary education	0.2488***	0.1707	0.4729***
Country*Tertiary education	0.1105	−0.4289	1.6100*
Country*Income level4	0.1040	0.2749	0.5462***
Country*Income level3	−0.2804	−0.0966	0.2331
Country*Income level2	−0.3713**	−0.3167	0.5384
Country*Income level1	−0.3750**	−0.4190***	0.1670
Country*Employment	0.0510	−0.0059	−0.2169
Constant	−2.3545*	−2.7886*	−3.9864*
Number of observation	6608	6608	6608
McFadden’s Pseudo R square	0.1136	0.1207	0.0931
Log Pseudo likelihood	−3937.6958	−3231.1113	−1358.9806
Wald χ²	828.61	734.21	232.26
Prob > χ²	0.000	0.000	0.000

Source: Calculated from World Bank’s Global Findex database (micro-data), 2017.

Note: *, ** and *** Significant at 1%, 5% and 10% level, respectively.

We observe that all the individual characteristics except gender have a significant impact on saving. Age has a non-linear relation with saving, with a positive coefficient for age and a negative coefficient for age square. It is clear that the probability of saving increases with age, but after a certain age saving diminishes. This is exactly what the lifecycle model claims. Therefore, our result supports the hump-shaped saving pattern as mentioned in Modigliani’s lifecycle model. Education is positively related to saving. The significant positive coefficient for both secondary and tertiary education, with a higher coefficient for the latter one, indicates that higher educated people are more likely to save. This is because people with higher education are usually expected to gain more income and thus save more money (Morisset et al., 1995). Moreover, education increases financial literacy, so that, there is a positive impact of education on savings. We have detected negative and significant coefficients for all income levels and coefficients become larger as we move towards lower-income quintiles. This specifies that greater income is associated with a higher probability of saving. The poor have low income that is almost spent on consumption, so that, they find it hard to save for the future. Employment is the only source of income and the ability to save will indeed be greater for income earners. Our finding is in line with the above prediction and shows that the likelihood of saving is higher for working people.

The estimated results provided in the above tables are sufficient to conduct a comparative assessment in terms of saving between India and China. The significant positive coefficient corresponding to the country dummy explains that Chinese adults are more probable to save rather than Indian adults. This outcome is not surprising given that per capita income (PCI) in China is dramatically higher than India’s PCI (World Bank, 2021). Interestingly, the coefficient associated with Country*Gender dummy is negative and indicates that being a female, saving probability decreases more in China. Having found such evidence, using the Blinder–Oaxaca decomposition technique (Sinning et al., 2008), we have tried to examine this result in greater detail. From the results reported in Table 3, it is very clear that, in terms of saving, both Chinese and Indian women are lagging behind their male counterparts. But for China, such inequality is considerably higher than in India. From Table 3, we can see that Chinese females are 9.26% less likely to save than their male counterparts, while, Indian females are 5.91% less likely to save compared with their male counterparts. This could be due to the higher socio-economic differences between males and females in China or it also can happen that, instead of saving, women in China prefer to spend more on consumption (China Briefing, 2012). This result needs further research. However, Chinese women are more likely to save as compared with Indian women (cross-tabulation reported in Tables A2 and A3). It is obvious because, in terms of employment, Chinese women are far ahead of Indian women (World Bank, 2021). Hence, they have more financial freedom too. We are now going to discuss the most conflicting issue in China, that is, the age-saving profile. The negative coefficient associated with the Country*Age dummy suggests that with age, saving probability increases more in India. The addition of the age coefficient (0.0639) and Country*Age dummy coefficient (–0.0992) present a more prominent picture to make an in-depth analysis. The resulting value (–0.0353) is simply the coefficient of age for China, which is negative in sign. Adopting a similar methodology, we can calculate the coefficient of age square for China, which would be positive in sign (0.0003). These findings show that, in China, with ageing, the probability of savings falls until a certain age after which the probability increases. This type of saving behaviour does not conform to the standard life-cycle model. However, this result is highly consistent with some macroeconomic studies (Chamon & Prasad, 2010; Lugauer et al., 2019), who have identified a U-shaped pattern of savings for China. This finding may be linked to the fact that young people in China save more, so that, they can create a buffer stock to invest their money in durables, such as a home (Yao et al., 2014), middle-aged people save less due to family expenses, that is, the cost associated with children including education and the unpredictable burden of healthcare expenditures leads the elderly to save more (Chamon & Prasad, 2010). Our research further shows that the impact of education and income on saving are much stronger in China than in India. Chinese educated adults are more interested in savings as compared with Indian educated adults (see Tables A2 and A3 for more details). However, the significant positive coefficient of the Country*Secondary education dummy explains, in terms of saving the gap between secondary-educated adults and primary or less educated adults is substantially higher in China than in India. Concerning income, the coefficients associated only with Country*Incomelevel 1 and Country*Incomelevel 2 dummies are significant with a negative sign. These results illustrate the fact, that income inequality has a profound impact on the saving likelihood in China than in India. The probability of saving among the poor is lower for both the countries (see Tables A2 and A3) as compared with rich adults. But, this savings gap in terms of different income groups is much more pronounced in China compared with India.

Table 3.

The Gender Gap in Saving, Modes of Saving

Country	The Blinder–Oaxaca Decomposition for Nonlinear Regression
Country	Saving	Formal Saving	Informal Saving
China	0.0926*	0.0780*	0.0084
India	0.0591*	0.0493*	−0.0202***

Source: Calculated from World Bank’s Global Findex database (micro-data), 2017.

Note: *, ** and *** Significant at 1%, 5% and 10% level, respectively.

Similar types of evidence also have been observed for both formal and informal savings. Individual characteristics, such as age, income, education and employment affect modes of saving in the same way that they affect savings. Age has a significant non-linear relation only with informal savings, illustrating that the likelihood of informal savings increases and then decreases after a certain age. Education positively influences formal saving but does not affect informal saving. It must be noted here that gender is significantly associated only with informal savings. Being a woman increases the probability of informal savings.

Chinese adults save more formally compared with Indian adults, but the converse is true when it comes to informal savings. The possibility of informal saving is higher in India than in China. However, the gender gap is higher for both modes of saving in China. In formal saving, the male–female gap for China is 0.0780 and for India is 0.0493 (Table 3). This positive gender gap suggests that males are more likely to save formally in both countries, but the gap is larger for China. Saving using an informal saving club is less in China. In this case, a significant gender difference (–0.0202) has been identified only for India. This result signifies the fact that females are 2.02% more likely to save informally than their male counterparts. The coefficients associated with Country*Age dummies suggest that with age, the probability of savings in financial institutions increases more in India as compared with China. Moreover, after adding coefficients corresponding to (a) age and Country*Age and (b) age square and Country*Age square, again an unusual (the opposite of the traditional model) U-shaped pattern of savings has been obtained for China. This result illustrates the fact that in China with age probability of formal saving decreases, but after a certain age, it again starts to increase. Furthermore, it has been discovered that with education and income likelihood of saving increases more in China than in India.

The main conclusion is that being older, richer, more educated and employed to a certain extent favour savings, either formally or informally, or both than others. Women resort more to informal savings than their counterparts. Age also has a non-linear relation with savings. Subsequently, the comparative analysis performed on saving behaviour between India and China made it clear that the likelihood of saving is higher among Chinese adults than the adults in India, but the reverse is true in the case of informal saving. However, the gap between male–female, higher educated-lower or uneducated and rich–poor in the case of saving is greater in China than in India. The unfamiliar U-shaped saving pattern with age also has been determined for China through our examination. This result needs further in-depth research mainly considering time-series data.

Socio-economic Factors Influencing the Various Motives Behind Saving: Broad Comparison Between India and China

A person always saves with a motive. Prior research on saving behaviour provides evidence of several saving motives, such as to smooth the level of consumption over the life (Modigliani & Cao, 2004), to overcome uncertain emergencies (precautionary motive), savings for education, health, business, retirement, etc. To make a more comprehensive study on savings, in this section, we investigate the determinants of two different motives behind saving: ‘for farm or business’ and ‘for old age’ and present a comparative analysis between India and China. Global Findex micro-data is available only for the two saving motives mentioned above for the year 2017, hence, we failed to capture the effect of socio-economic factors on other saving motives. This is the main constraint of our research.

Table 4 displays the estimated results. Other than gender, all the individual characteristics have a considerable effect on saving motives. Age has the same relationship with both the motives of saving. The relation is non-linear. We can observe that with age the probability of saving for any purpose increases and then decreases after a certain age. Dummy variables for income are all significantly negative for each saving motive with larger coefficients for lower-income quintile dummies, illustrating the fact that being poor decreases the probability of saving for any motivation. Education and employment are positively related to each saving motivation; suggest that being educated or employed increases the likelihood of saving for both old age and farm/business.

Table 4.

Determinants of Various Saving Motives

	Pooled Logistic Regression
	Savings for Old Age	Savings for Farm/Business Purposes
Country	1.5903**	2.0882*
Gender	0.0907	−0.1180
Age	0.0969*	0.0668*
Age square	−0.0007*	−0.0006**
Secondary education	0.7104*	0.5472*
Tertiary education	1.0491*	0.9664*
Income level 4	−0.4077**	−0.2755
Income level 3	−0.4807*	−0.0127
Income level 2	−0.8675*	−0.3795***
Income level 1	−1.3809*	−0.7756*
Employment	0.6636*	0.7798*
Country*Gender	−0.3203**	−0.2502
Country*Age	−0.0262	−0.0326
Country*Age square	0.0003	−0.0002
Country*Secondary education	−0.1554	−0.1917
Country*Tertiary education	−0.3210	−0.3777
Country*Income level 4	0.1894	0.3719
Country*Income level 3	−0.1635	−0.1414
Country*Income level 2	0.0335	−0.2628
Country*Income level 1	0.0294	0.0653
Country*Employment	−0.2322	0.0427
Constant	−4.8726*	−4.2337*
Number of observation	6608	6608
McFadden’s pseudo R square	0.0969	0.0982
Log pseudo likelihood	−2718.7519	−2034.8849
Wald χ²	449.05	351.85
Prob > χ²	0.000	0.000

Source: Calculated from World Bank’s Global Findex database (micro-data), 2017.

Note: *, ** and *** Significant at 1%, 5% and 10% level, respectively.

Now, we will come to our main question which country is performing better in respect of saving for these two purposes. China is ahead of India. Being Chinese increases the probability of saving for old age and business compared with India. Here, the coefficient of Country*Gender dummy is significant only for old age saving, which explains that gender difference in old age saving is higher in China. Table 5 provides a more prominent picture in this aspect. Results show that Chinese females are 6.34% less likely to save for old age than their male counterparts, while this result is 3.04% in the case of India. Conversely, in the case of saving for farms/businesses, the gender gap (0.0427) is marginally higher in India as compared with China (0.0403). The higher economic opportunity of women in China might be the possible explanation for this result. The score of women business and the law index is also higher in China (World Bank, 2021), which increases the tendency among Chinese women to save for business. All other coefficients corresponding to interaction dummies are highly insignificant; therefore, there is no difference between China and India in terms of these variables.

Table 5.

The Gender Gap in Various Saving Motives

Country	The Blinder–Oaxaca decomposition for nonlinear regression
Country	Savings for Old Age	Savings for Farm/ Business Purposes
China	0.0634*	0.0403*
India	0.0304**	0.0427*

Source: Calculated from World Bank’s Global Findex database (micro-data), 2017.

Note: *, ** and *** Significant at 1%, 5% and 10% level, respectively.

V. Conclusion and Policy Implication

At the individual level, saving works as a cushion against unforeseen financial emergencies, particularly for the poor. It also provides support at old age. On the other hand, in a wider context, savings is essential for economic growth. Our study has made a comprehensive analysis of saving behaviour using the micro-data of India and China. The primary reason to conduct this research is to understand the role of some socio-economic variables, such as gender, age, income, education, employment in determining the saving behaviour and make an extensive discussion on the saving performance of China and India. The findings and inferences of this article are based on the econometric tool, that is, pooled logistic regression model, which is used to carry out our analysis and the Blinder–Oaxaca decomposition technique has been used for estimating the gender gap in terms of saving in two countries. Estimated results confirm that age, income, education, employment are the most important determinants of saving including formal and informal saving and saving motives. We have identified that being richer, educated, employed and older favour saving, both formal and informal method of saving, savings for old age and farm than others. Age also has a non-linear relation with saving, saving methods and motives, which supports the view of Modigliani in respect of age-saving patterns for pooled data. The significant impact of gender only has been observed for informal saving, which illustrates that women are more prone to save informally than men.

This article also compared China with India. Evidence suggests that China is ahead of India in terms of saving, formal saving and saving motives, while, informal saving is far more common in India than in China. The gender gap in saving, formal saving and saving for old age is higher in China than in India. The reverse is true in the case of saving for farms or businesses. Furthermore, it has been observed that the effect of income and education on saving is much stronger in China as compared with India, which establishes that with education and income probability of saving increases more in China. Therefore, inequality in saving behaviour between (1) poor and rich, and (2) higher educated and lower or uneducated is also larger in China than in India. These results need further examination to understand the economic as well as social conditions of women and the poor in these two countries. Are the government policies, public institutions more favourable in India for the poor compared with China? Does Jan DhanYojana, MNREGA, etc. are responsible for the lower gap between rich and poor in India? Despite having a higher literacy rate, employment and financial independence among Chinese women, why the gender gap in saving is higher in China? In India microfinance, self-help groups are very proactive and have instilled saving behaviours among Indian women, particularly poor Indian women, which can be a reason behind the lower gender gap in India. This result also needs further research. It is worthwhile to mention here that, likewise Chamon and Prasad (2010) and Lugauer et al. (2019), our study also has found evidence in favour of a U-shaped age-saving pattern, which is not consistent with the predictions of the standard version of life cycle hypothesis. This result also needs further in-depth analysis considering time-series data.

Based on the findings of this article, we make the following policy prescriptions. First, the contribution of income highlights the fact higher income leads to a higher probability to save, both formally and informally. Similar evidence manifested also in the case of saving for old age and business/farm. Therefore, unless and until people have a sufficient amount of income, their interest in savings will not grow. Second, the influence of employment can be considered as supportive evidence of the above. Third, education, particularly secondary education and above, should continue to be a national priority. To sum it up, policies should be designed with these aspects in mind, so that, people can get higher education, get job opportunities and earn a decent income.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Appendix A

Notes

References

Akram

(2021). Household’s saving behaviour in Pakistan: A micro-data analysis. Studies in Microeconomics. 10.1177/23210222211024377

Ang

(2009). Household saving behaviour in an extended life cycle model: A comparative study of China and India. The Journal of Development Studies, 45(8), 1344–1359. 10.1080/00220380902935840

Ariç

K. H.

(2015). Determinants of savings in the Middle East countries. KOSBED, 29, 23–36.

Bohini

, Yaganti

C. H.

, & Thomas

M. P.

(2021). India’s demographic dividend: The Millennials’ savings conundrum. Millennial Asia. 10.1177/0976399620958262

Chamon

M. D.

, & Prasad

E. S.

(2010). Why are saving rates of urban households in China rising? American Economic Journal: Macroeconomics, 2(1), 93–130. 10.1257/mac.2.1.93

China Briefing. (2012). Consumption trends and targeting China’s female consumers. https://www.china-briefing.com/news/consumption-trends-and-targeting-chinas-female-consumer/

Clancy

, Grinstein-Weiss

, & Schreiner

(2001). Financial education and savings outcomes in individual development accounts (HEW 0108001). University Library of Munich.

De Vos

, Obokoh

L. O.

, & Abiola

B. A.

(2020). Determinants of savings among non-Ricardian households in South Africa. International Journal of Social Economics, 47(11), 1329–1343. 10.1108/IJSE-11-2019-0692

Demirgüç-Kunt

, Klapper

, Singer

, Ansar

, & Hess

(2018). The Global Findex Database 2017: Measuring financial inclusion and the Fintech revolution. World Bank. https://microdata.worldbank.org/index.php/catalog/global-findex

10.

Domar

E. D.

(1946). Capital expansion, rate of growth, and employment. Econometrica, 14, 137–147. https://www-jstor-org-443.web.bisu.edu.cn/stable/1905364

11.

Ghosh

, & Hom Chaudhury

(2019). Gender gap in case of financial inclusion: An empirical analysis in Indian context. Economics Bulletin, 39, 2615–2630.

12.

Gokhale

(2000). Are we saving enough? Economic commentary. Federal Reserve Bank of Cleveland.

13.

Harrod

(1939). An essay in dynamic theory. Economic Journal, 49, 14–33. https://www-jstor-org.web.bisu.edu.cn/stable/2225181

14.

Horioka

C. Y.

, & Wan

(2007). The determinants of household saving in China: A dynamic panel analysis of provincial data. Journal of Money, Credit and Banking, 39(8), 2077–2096. 10.1111/j.1538-4616.2007.00099.x

15.

Hua

T. X.

, & Erreygers

(2019). Applying quantile regression to determine the effects of household characteristics on household saving rates in Vietnam. Journal of Asian Business and Economic Studies, 27(2), 175–193. 10.1108/JABES-06-2019-0053

16.

Hung

J. H.

, & Qian

(2010). Why is China’s saving rate so high? A comparative study of cross-country panel data (Working Paper Series). Congressional Budget Office.

17.

Lean

H. H.

, & Song

(2009). The domestic savings and economic growth relationship in China. Journal of Chinese Economic and Foreign Trade Studies, 2(1), 5–17.

18.

Lee

, & Pocock

(2007). Intrahousehold allocation of financial resources: Evidence from South Korean individual bank accounts. Review of Economics of the Household, 5, 41–58. 10.1007/s11150-007-9004-3

19.

Lugauer

, Ni

, & Yin

(2019). Chinese household saving and dependent children: Theory and evidence. China Economic Review, 57(C), 101091. https://dx-doi-org.web.bisu.edu.cn/10.1016/j.chieco.2017.08.005

20.

Modigliani

(1986). Life cycle, individual thrift and the wealth of nations. American Economic Review, 76, 297–313. https://www-jstor-org-443.web.bisu.edu.cn/stable/1813352

21.

Modigliani

, & Cao

S. L.

(2004). The Chinese saving puzzle and the life-cycle hypothesis. Journal of Economic Literature, 42(1), 145–170. 10.1257/002205104773558074

22.

Morisset

, Revoredo

, & Division

(1995). Savings and education: A life-cycle model applied to a panel of 74 countries (Working Paper No. 1504). World Bank Policy Research. 10.1596/1813-9450-1504

23.

Nwosu

, Anumudu

, & Nnamchi

(2019). Microeconomic determinants of household savings in Nigeria. Journal of International Development, 32(2), 150–167. 10.1002/jid.3440

24.

Samantaraya

, & Patra

(2014). Determinants of household savings in India: An empirical analysis using ARDL approach. Economics Research International, 2014, 454675, 10.1155/2014/454675

25.

Schultz

T. P.

(2005). Demographic determinants of savings: Estimating and interpreting the aggregate Association in Asia (Discussion Papers, No.1479). Institute for the Study of Labor (IZA).

26.

Seguino

, & Floro

(2003). Does gender have any effect on aggregate saving? An empirical analysis. International Review of Applied Economics, 17, 147–166. 10.1080/0269217032000064026

27.

Sinning

, Hahn

, & Bauer

(2008). The Blinder–Oaxaca decomposition for nonlinear regression models. The Stata Journal: Promoting Communications on Statistics and Stata, 8(4), 480–492. 10.1177/1536867X0800800402

28.

Solow

R. M.

(1956). A contribution to the theory of economic growth. Quarterly Journal of Economics, 70, 65–94.

29.

Tobing

(2012). Demography and cross-country differences in savings rates: A new approach and evidence. Journal of Population Economics, 25(3), 963–987. 10.1007/s00148-011-0386-4

30.

Wooldridge

J. M.

, (2012). Introductory econometrics: A modern approach. South-Western Cengage Learning.

31.

World Bank. (2021). World development indicators. The World Bank. https://databank.worldbank.org/source/world-development-indicators

32.

Yao

, Xiao

J. J.

, & Liao

(2014). Effects of age on saving motives of Chinese urban consumers. Journal of Family and Economic Issues, 36, 224–238. 10.1007/s10834-014-9395-2

33.

Zhang

, Hang

, & Liu

(2016). Multi-scale causality between saving and growth: Evidence from China. Applied Economics Letters, 24(11), 790–794. 10.1080/13504851.2016.1229403