Introduction

In the world, India has the largest diaspora, with a diversified global presence of about 17 million [1]. It is observed that the North American region consists of nearly 3 million, while six Gulf Cooperation Council (GCC) countries contain 9 million emigration stock [1]. However, it found from the literature that the Gulf emigration of low-skilled workers from India is due to the distress factors such as relatively higher earnings quest, unemployment and poverty [2,3,4,5]. Moreover, large-scale economic activities due to the discovery of oil in the Gulf region create employment opportunity for low-skilled workers [6, 7]. Although the emigration to the Gulf region has already been well established in the literature, the emigration of relatively skilled employees from India to the United States of America (USA) is yet to explore. Thus, it is necessary to find out the determinant or motives of emigration from India to the USA.

A variety of economic, social, demographic and conditions in the destination and origin countries motivate the potential migrants to leave their origin country. It is true that migration tendency takes place when the mentioned conditions deteriorate in the origin country (i.e., push factors) or these factors improve in the destination country (i.e., pull factors). After calculating the socio and economic costs and benefits, the potential migrants take the decision to migrate, provided the benefits of migration exceed the costs of migration [8]. Thus, understanding the chemistry of what motivates the migration is necessary to frame a policy for discouraging the brain drain and encouraging brain gain.

In this vein, economic and non-economic factors that determine migration have explored the economic research for decades. However, not yet fetched a uniform conclusion about the determinants. Hence, understanding the determinants would help the policymakers to make domestic policies for migration to have a direct or indirect impact. Further, the implications of migration can be drawn by policymakers. Therefore, socio-economic and political factors have been explored in the literature [9,10,11].

Moreover, the present focus migration studies focus on micro-level aspects and panel-level studies [2, 12, 13]. Thus, it is unable to fetch adequate conclusions for formulating country-specific policies at the macro level. Further, the existing studies in the literature explore the determinants of low-skilled workers' emigration from India [2, 3]. However, no study has been conducted to identify determinants of high-skilled workers' emigration from India. Hence, focusing on such determinants helps us to discourage the brain drain and encourages brain gain.

As a result, this study considers the USA as the destination of high-skilled workers from India due to many reasons. First, in the early nineteenth century, Indian immigrants began arriving in the United States and settling in communities along the West Coast. Despite the fact that they arrived in small numbers initially, new possibilities arose in the mid-twentieth century, and the population grew in the decades that followed. As of 2019, there were over 2.7 million Indian immigrants in the United States [14]. Second, Indian immigrants presently account for around 6% of the foreign-born population in the United States, making them the country's second-largest immigrant group, ahead of Chinese and Filipino immigrants. The Indian immigrant population in the United States surged 13-fold between 1980 and 2019. Indian immigrants are well-known for their diversity as well as their large numbers. Indian immigrants are more likely than the general U.S. and foreign-born populations to be well educated, work in management positions, and be businessmen, IT and health professionals [2]. Third, according to the World Bank report, the United States sent USD 68 billion in inward remittances to India in 2020. All of these underline the significance of high-skilled workers’ emigration to the USA from India in the recent past.

Given the background of this study, attempts have been made to examine the economic, social and demographic determinants of emigration from India to the US. Understanding the factors that influence emigration is crucial for two reasons. First, it aids immigrants' social and economic assimilation into their host countries. Understanding what drives high-skilled emigrants to leave their home nations allows us to better understand their motivations for emigrating and better accommodate those motivations in host countries. Second, examining the factors that influence movement increases the effectiveness of migration policy.

The rest arrangement of the study is as follows. Section 2 offers a review of related literature. Section 3 contains the model, data and econometric procedure. Section 4 contains results and discussion. The final section concludes with policy suggestions.

Review of related literature

There is a growing body of knowledge about the links between human migration, unemployment, remittances, and the impact of migration on migrants and their families [2,3,4,5]. However, the relationship between emigration, GDP growth, human development, and population density, on the other hand, is poorly understood. An investigation into this relationship is necessary because policymakers in India are concerned about the impact of population density, human development, and GDP growth in their home countries on the emigration of high-skilled workers. According to economic theory, increased income in underdeveloped countries will reduce emigration. However, it found that emigration grows to a limited extent as the economy improves and decreases as the economy reaches the upper–middle income level (Clemens 2014). Hence, migration is a cost to move from backward economic circumstances to a forward one [15,16,17]. As per capita income in the home country rises, migration will likely decline [18, 19]. At lower income levels, emigration rises as income rises, but after a turning point, emigration falls as income rises. So there is an inverted “U”-shaped relation between income and migration. It has been known under different headings, namely “mobility transition” [20], and “migration curve” [21].

Further, it was observed that workers from developing countries usually choose to work in countries with greater possibilities. Hence, there is an inverse relationship between a country’s economic prosperity and its population emigration [22]. It also found that people from the wealthiest countries are more likely to migrate than those from the poorest [23]. Another viewpoint is that emigration trends differ from one country to another. Both educated and unskilled people are less likely to emigrate if they can earn more money in their home country. Educated people are more likely to leave if their earnings in a foreign country are higher than their own. Hence, domestic economic growth not only raises earnings for both categories, but also moves people from lower to higher education levels [24].

The relationship between human development and migration was investigated by Leonie Decrinis [25]. According to his research, the relationship follows an inverted U-shaped curve. Because of this connection, many politicians in Western countries have had to implement a variety of socioeconomic development policies in order to slow the rate of emigration. However, evidence suggests that such migration-reduction tactics are ineffective. Between 1990 and 2000, Lucas [26] conducted an empirical study on migratory flows from Sub-Saharan African countries to OECD destinations. The study discovered that countries with per capita incomes exceeding 1.000 US dollars have a higher rate of cross-border migration than those with incomes below this level.

Lazareva [27] looked into the primary factors that influence migrants' destination choices. It identified the key factors of destination choice as the expected lifetime income in the event of finding a job, the likelihood of getting a job, and the cost of moving and living in the destination. Further, Gallaway and Vedder [28] investigated the push and pull factors that influence migration between the United Kingdom and the United States. The main push factors for migration are the low wage and high unemployment in the United Kingdom and the relatively high wage and low unemployment in the United States. People move for a variety of reasons, including economic considerations such as income and employment disparities, as well as the existence of infrastructural improvements in large cities, such as enhanced education and health facilities [29]. According to the UNDP's Global Human Development Report [30, 31], three fourth of international migrants move to nations with better human development than their home countries.

It was found that many scholars conducted a variety of migration studies over a period of time. Some were concerned with the impact of human development on emigration, while others were concerned with the impact of economic growth on human mobility. However, there is no unanimous conclusion on the relationship. Moreover, there is no research which has been done on the impact of India's GDP growth, human development, and population density on emigration to the United States. Moreover, this research utilizes newly updated data on GDP growth, India's human development index, population density, and emigration from India to the United States (1995–2019). Furthermore, the use of advanced time series techniques employment for the empirical analysis distinguishes this work.

The model, data and econometric procedure

There are certain factors (GDP growth, human capital formation, population density) responsible for the emigration of a human beings. Here we analyze the effect of change in all these factors on emigration from India to the USA. We construct an econometric model to examine the link between emigration to the US and its various determinants in Eq. 1.

$${\text{EMIUS}}_t = \beta_0 + \beta_1 {\text{GDP}}_t + \beta_2 {\text{HC}}_t + \beta_3 {\text{DP}}_t + \mu_t$$
(1)

where, EMIUS is the emigration to the United States in different periods, GDP is the growth of GDP of India, HC measures the human capital formation, DP means the density of the population, μ is an error term.

The given model is estimated by using the data from 1995 to 2019. The data on emigration from India to the US are converted to a yearly basis (from five-year by using the quadratic match sum method as the method adjusts for seasonality [32]. The data were collected from various sources such as the World Bank 2020, Penn world data for human capital, and World Population Prospects (United Nations) for emigration.

In order to estimate the model, we use the Auto Regressive Distributed Lag (ARDL) approach as the use of ARDL eliminates the problem of endogeneity in research by accommodating the lags of dependent and independent variables [33]. The main advantage of using ARDL is that it can be used irrespective of whether the regressors are I (1) or I (0), and thus, it eliminates the pre-testing problem for unit root [34]. Hence, it assesses the long-run relationship between the variables by using the unrestricted error correction model (UECM) as reported in Eq. 2.

$$\Delta {\mathrm{LNEMIUS}}_{t} = \,{\theta }_{1}+{\theta }_{2}{\mathrm{LNEMIUS}}_{t-1}+{\theta }_{3}{\mathrm{LNGDP}}_{t-1}+{\theta }_{4}{\mathrm{LNHC}}_{t-1}+{\theta }_{5}{\mathrm{LNDP}}_{t-1}+\sum_{i=1}^{p}{\alpha }_{2i}{\Delta \mathrm{LNEMIUS}}_{t-i}+\sum_{i=0}^{m}{\alpha }_{3i}{\Delta \mathrm{LNGDP}}_{t-i}+\sum_{i=0}^{n}{\alpha }_{4i}{\Delta \mathrm{LNHC}}_{t-i}+\sum_{i=0}^{k}{\alpha }_{5i}{\Delta \mathrm{LND}P}_{t-i}+{\varepsilon }_{1t}$$
(2)

Here \({\varepsilon }_{1t}\) symbolizes the error term by satisfying the normal distribution with constant variance and zero mean. Further, the lag operator and the constant are represented by using \(\Delta\) and \({\theta }_{1}\), respectively. The long-run coefficient of the independent variable on the dependent variable is outlined from \({\theta }_{2}\) to \({\theta }_{5}\). However, the long-run relationship among the series can be established by using the F-test. In this line, a hypothesis test, i.e., H0 \({: \theta }_{2}\)=\({\theta }_{3}\)=\({\theta }_{4}={\theta }_{5}=0\) vs H1:\({\theta }_{2}\ne {\theta }_{3}\ne {\theta }_{4}\ne {\theta }_{5}\ne 0\) has been conducted. A set of two critical values proposed by [35] has been used for comparing the estimated F-statistics, in which 1st difference is denoted by I (1) or upper bound and the level signifies I (0) or lower bound. Based on this, the alternative hypothesis is accepted, or the null hypothesis is rejected if the upper bound value is below the analyzed value of the F-statistics. Similarly, the rejection of the alternative hypothesis happens when the analyzed F-statistics are below the lower bound. In the same test, the inclusive results would be possible when the F-statistics fall in between upper and lower bounds.

Further, the approaches use the Error Correction Model and which is derived from the ARDL model through a simple linear transformation as Error Correction Model (ECM) integrates short-period adjustments with long-period equilibrium without losing long-period information and is represented in Eq. 3.

$$\Delta {LNEMIUS}_{t}\,=\,{\theta }_{1}+\sum_{i=1}^{p}{\alpha }_{2i}{\Delta LNEMIUS}_{t-i}+\sum_{i=0}^{m}{\alpha }_{3i}{\Delta LNGDP}_{t-i}+\sum_{i=0}^{n}{\alpha }_{4i}{\Delta LNHC}_{t-i}+\sum_{i=0}^{k}{\alpha }_{5i}{\Delta LNDP}_{t-i}+{\gamma }_{1}{ECM}_{t}+{\varepsilon }_{1t}$$
(3)

Here, \(\gamma\) shows that in a short period of time, how the variables run to the long-run association. Toward confirming this, it is the hypothesis that there is a negative coefficient sign and less than 0.5% for the error correction term (ECT). Further, the cumulative sum (CUSUM) and the cumulative sum of squares (CUSUMQ) have been used for measuring short-run stability along with heteroscedasticity, normality, and casual correlation to confirm the model’s goodness of fit.

Results and discussion

The descriptive statistics have been carried out in the first analysis phase. Thus, In Table 1, the result of descriptive statistics is portrayed, in which the higher average value is for EMIUS, followed by GDP, DP and HC, respectively. In terms of volatility, the variables follow a similar path. However, EMIUS, DP and HC have the higher left tail while GDP has the right tail. Moreover, the variables follow a normal distribution as the significance of the Jarque–Bera test is far from the 5% and 1% levels of statistical significance.

Table 1 Descriptive statistics
Full size table

To examine the long-run relationship among EMIUS, GDP, HC, and DP, the study checks for the stationarity of each variable in the first stage. It believes that if the variables in the model are not integrated, then the standard inferences can be obtained. Thus, the stationary properties of the time-series observations have been conducted by using standard unit-root tests. In order to do this, the Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) approaches have been used as it controls for serial correlation in the unit root estimation process. The outcomes of the tests in Table 2 reveal that the rejecting null hypothesis of unit roots is impossible at the levels of all variables. But, the conventional significance levels have been observed for the first difference of the variables across the tests. Hence, it found that all variables are I (1) series.

Table 2 Unit root test results
Full size table

After ascertaining the integration order of the variables, the long-run relationship among the variables in the US emigration function has been estimated within the ARDL framework. However, the selection of appropriate lag length is important since the ARDL model is more sensitive to lag. Thus, it uses the Akaike information criterion (AIC) for opting for appropriate lag. Besides, due to appropriateness for a smaller number of observations, the study considers Narayan’s (2005) critical upper [I (1)] and lower [I (0)] bounds statistics as a benchmark for comparing the estimated F-statistics. In Table 3, the reported ARDL cointegration test result indicates that estimated F-statistics are above the upper bound value of Narayan (2005). Therefore, the long-run relationship is confirmed among the study variables.

Table 3 The results of ARDL cointegration test
Full size table

Further, the ARDL model-based estimates for long-run and short-run are reported in Table 4. The outcome indicates that emigration to the USA is negatively associated with India’s economic growth. Precisely, a 1% increase in economic growth in India leads to a reduction in the emigration of about − 0.226%(− 0.376) in the long run (short-run). It means that if there is an increase in prosperity and productivity of the economy, then the people are less likely to emigrate as they are getting a better standard of living in the home country itself. Thus, there is no need to inquire about overseas economic opportunities for improving the living standard. This finding gives a powerful insight into the need for billions of dollars of investment for promoting economic development by checking the “root causes of migration”.

Table 4 Long and short runs result estimates
Full size table

Besides, it found that emigration to the USA is positively linked with India’s human capital development. Specifically, a 1% increase in human capital development promotes emigration by 2.04% (0.761) in the long run (short-run). This finding implies that a skilled workforce and innovative minds sets are going out from India because of decent livelihood opportunities. Further, it is attributed to the lack of avenues for higher education after attaining the necessary skills. Thus, they move to other countries to attain sufficient skills in their fields, thereby finding a good opportunity for a decent life. Additionally, the undermining of the talent even after attaining skills is another possible cause of emigration.

Additionally, this study reveals that population density positively impacts emigration to the USA. In precise, a 1% increase in population density enhances the emigration by 3.87% (1.45%) in the long run (short-run). The possible cause for such a finding could be that increased population in a particular region necessitates the movement of people from one region to another to find better job opportunities, thereby having a decent lifestyle. Another possible reason could be the imbalances in the development of the country. However, the imbalances in economic growth create unequal distribution in densely populated areas. Thus, the people from the area move to other countries to have better income, thereby improving their living standards.

Finally, the study conducted the square of recursive residuals (CUSUMsq), and the cumulative sum of recursive residuals (CUSUM) has been estimated and plotted in Fig. 1. It unveils the stability of the parameters of the EMIUS model for all the periods except 2008. This structural instability may be due to the global financial crisis that created structural change emigration to the USA. Further, the heteroscedasticity, stability as well as the serial correlation of the ARDL model have been tested. It founds that there is no such issue presented in the model (See Table 4). Further, the coefficient of determination (R2 and Adjusted R2) shows that about 92% variation in the emigration to the US is determined by these factors. Besides, the non-existence of autocorrelation can be observed from the Durbin–Watson stat as it is on par with the standard value.

Fig. 1
figure 1

CUSUM and CUSUMsq tests at 5% level of significance

Full size image

Conclusion and policy implications

Global migration is one of the world’s most challenging present and future concerns. While exporting countries grapple with the negative consequences of brain drain, receiving countries face the problem of integrating migrants of all cultures and nationalities into their domestic labor markets and societies [36]. In this vein, this study examines the relationship between emigration from India to the United States with India's GDP growth, human capital, and population density. The study of such a link is critical, as policymakers in India are increasingly concerned about skilled worker brain drain. It is claimed that higher rates of emigration from India to industrialized countries are linked to higher salaries, better job prospects, and education [2]. The findings of this study reveal how economic growth, human capital, and population density all play a role in predicting emigration to the United States.

Thus, the empirical findings can be concluded as follows. The finding emanating from the study confirms the long-run relationship between the variables. Besides, the result shows that emigration to the United States has a detrimental impact on India's economic growth. Further, it discovered that emigration to the United States increases as an increase in India's human capital grows. Furthermore, this research shows that population density has a positive impact on emigration to the United States.

The ramifications of these findings for policy are important. The negative relationship between India's GDP growth and emigration to the United States reveals the urgent need for billions of dollars in investment to promote economic development by addressing the “root causes of migration”. Emigration to the United States and the building of human capital in India are intrinsically linked. This research shows that India's qualified labor and innovative brains are leaving due to a lack of suitable employment opportunities. It was also ascribed to a lack of opportunities for additional education after obtaining the required abilities. As a result, they migrate to other nations in order to gain appropriate abilities in their industries and, as a result, to find a good opportunity for a comfortable living. This type of movement of brilliant workers is a form of “brain drain”, and the authorities must take the appropriate actions to stop it and use these workers in the growth of our country. Finally, there is a positive relationship between population density and emigration to the United States. It highlights the country's development inequalities. However, in densely populated places, economic development imbalances result in unequal distribution. As a result, residents of the area migrate to nearby countries in search of a higher level of living. So the government should take the required steps to promote “balanced growth”.