Return migration and development in rural communities : The case of Nzega and Magu districts , Tanzania

This study examines the effects of both internal and cross-border re-migration on rural development in Tanzania mainland. Four rural communities drawn from Nzega and Magu districts were selected based on a high magnitude of rural-to-urban and cross-border labour migration; and distance from the district headquarters. 648 re-migrants were sampled from the four communities. The theoretical framework applied in this study is a structural approach to re-migration. The analysis draws on data collected by using a descriptive cross-sectional design, adopting a mixed methods approach and a Probit regression technique used to estimate the effects of total domestic and overseas savings on various expenditure categories. Results confirm the overall hypothesis that return migration has an influence on rural development in the study areas. The study recommends for policy makers to improve the resource capacity of current migrants and their beneficiaries by stimulating rural economies through rural enterprise activities.


INTRODUCTION
Return migration has become an important livelihood strategy for many poor people across the world.Population mobility is a prerequisite condition for sustainable rural development and poverty alleviation in Africa.In the recent past, it has received increased attention from destination and origin countries due to both groups being interested in leveraging return migration to their economic advantage.Previous research on remigration has come out with three basic channels through which return migrants can facilitate rural development of their home countries.The first channel is through their garnered financial capital which can be invested upon return (Piracha and Vadean, 2010;De Vreyer et al., 2010) and invigorate the transfer or adoption of new technologies (Plaza and Ratha, 2011).The acquired knowledge, skills and familiarity with recent technologies which can be utilised upon return provide a second channel (Mahuteau and Tani, 2011;Kuschminder and Butcher, 2012).Thirdly, the augmented social capital of return emigrants can stimulate rural development through establishing new networks, perspectives as well as norms and values (Castaldo et al., 2012).

An overview on labour migration in Africa
Migration is a routine livelihood strategy of poor households which helps to smooth seasonal income fluctuations and earn additional money to meet contingencies or increase disposable income.It is one of many activities in the diversification portfolio of households and is undertaken if and when there is surplus labour in the household and when it can be combined successfully with other activities either in parallel or in sequence.Migration in Africa is driven by the often interrelated push factors of poverty, mismanaged economies, unemployment, unstable politics, insecurity and violent conflicts, rapidly growing populations as well as environmental deterioration.The remarkable economic growth rate recorded by many African countries in the last decade has been adversely affected by falling commodity prices and the weaker global economy; and besides the challenge to make growth more inclusive especially for the young population, it has not helped uplift poor people from poverty.Poor living conditions have rendered real wages well below the requirements for mere survival.In many of these countries, living conditions are even worse now than they were at independence in the 1960s and early 1970s.Unemployment across the continent ranges from 18 to 50%, mostly among youths, who are willing to do any work, anywhere and at any wage (Adepoju, 2010).
Africa is a continent of contradictions: though it is rich in resources, the sheer numbers of Africans living in poverty have grown, a fact reflected in the rapid growth of Africa"s population (World Bank, 2015a).Emigration from the continent takes two forms: firstly, emigration of skilled professionals and, secondly, emigration of intermediateskilled, semi-skilled and unskilled persons (Adepoju, 2015b).The last of these groups in particular is composed of both regular and irregular migrants.Within the continent, a large portion of the migrants from Uganda, Kenya, Ghana and Nigeria to Southern Africa, especially to Namibia, Botswana and the Republic of South Africa, consist of professors, engineers, accountants, pharmacists and kindred professionals who help staff the universities and other tertiary institutions of the destination countries (Adepoju, 2010;Oucho, 2015).These skilled workers have usually chosen to move because of political and economic crises in their home countries, as well as inadequate wages and a poor working environment.

Migration in Tanzania
Similar to other countries in the region, there is a long history of internal migration in Tanzania, dating to the colonial and especially pre-colonial times.The nation has witnessed significant labour recruitments historically from the neighbouring areas to provide a workforce for Masanja 73 plantation agriculture.Socialism development cut down the rural to rural distance migration owing to policy focus on community level farming.However, seasonal labour migration occurred in rustic areas as extra labour required during the harvesting, counter example is to where there were communal sisal farms near the cost.Migration in Tanzania continued to increase during the government decentralization and creation of new regional capitals in the1970s.The economic liberation in the1980s accelerated the rural to urban migration, which became permanent with migrants investing profits in venture in or near the town rather than remitting to their areas of origin.
Historically, there was little emigration of Tanzanians towards the western world.Currently, there are clear indications that there is a soaring number of Tanzanian migrant workers, both women and men working abroad.Tanzania is experiencing significant out-migration of the youths from low productivity agriculture to urban informal service sectors, where productivity is just as low.Rural areas in Tanzania are characterized by many factors which trigger massive population growth in urban areas, some of the rural characteristics include lack of employment opportunities, lack of business opportunities, poor social services, poor production, low income, lack of access to market, low price of agriculture output and high price of agriculture inputs and economic hardship.The movement is composed of legal and illegal as well as skilled and unskilled migrants.A significant proportion of the skilled Tanzanians migrate to more developed economies particularly in North America and Western Europe.However, it is important to state here that migration from Tanzania to Europe is a rather recent phenomenon, taking off mainly in the1990s and 2000s (UNHCR and IOM, 2010).
A significant number of Tanzanians also proceed to SADC countries mainly Botswana, South Africa and Swaziland.Most of the unskilled Tanzanians migrate to South Africa.Other research reports-such as that of Delius (2017) underscore the point that most Tanzanians who move to countries within the South African Migration System originate from rural localities.In Tanzania almost double as many leave the country than enter in comparison with the sub-Saharan Africa average.This is frequently due to native Tanzanians migrating for labour opportunities.

Return migration in Tanzania
Given their strong attachment to home areas and transient residence at the career destination, whether rural or urban, African workers tend to visit or migrate back to their homes after a period away.Research reports on re-migration to Tanzania are inadequate in the literature.However, the dearth of such literature carries scanty information which explains that there have been some movement streams returning from the well-endowed countries of the SADC region to the less endowed rural localities of Tanzania (Thomas and Inkpen, 2013).
More recent reports indicate that there is now an increasing trend of migrants returning home from overseas notably from European countries (Avato et al., 2010).The apparent shortfall is that available literature on returnees to Tanzania does not carry sufficient information pertaining to their economic well being.

The potential of return migrants
It is often pointed out that returnees can make a substantial contribution to rural development in the sending countries not only by investing the money they accumulated in their host country but also by stimulating the transfer of technological and managerial know-how.High economic growth rates can in turn stimulate return migration, leading to a virtuous cycle to the benefit of the sending country.Thus, next to remittances, the return of high skilled migrants is often regarded as one of the primary means of making migration work for economic development in the sending countries.
Although there is now a growing trend of returnees investing in neighbouring urban centres rather than their native villages in Tanzania as documented by Mbonile (2002), Fagerlund (2010) and Msigwa and Mbongo (2013) and that this return is in a substantial way linked to unsuccessful migration as Hirvonen and Lilleør (2015) have unpacked in their research report which used a 19 year tracking data; there is a substantial proportion of returnees investing in their places of origin and this paper makes reference and uses them to argue in favour of investments in native villages.Additionally, it is claimed that those who out-migrated do not seem to be more outstanding in entrepreneurship than those who remained at home (Hirvonen and Lilleør, 2015).However, systematic and rigorous reviews in the literature on international migration, provide that migrants who come back home appear to more business-oriented contrary to those who never moved out of their home communities.
Several reviewed empirical analyses have concentrated on the labour market performance and entrepreneurial skills of international migrants after their return, and these studies have generally found that return migrants outperform non-migrants; either due to selection or to skills reaped during the period of migration (De Vreyer et al., 2010).
Several researches that have investigated the capacity of migration to have an effect on the use of agricultural skills, methods, and processes of rural households have also delivered conflicting perspectives.The limited availability of labour emanating from rural out-migration presented by the pessimistic stance underscores the point that this limitation gives rise to the breakdown and deserting of the long established agricultural technologies which require a large workforce by households of migrants (Zimmerer, 2014;Garcia-Barrios and Garcia-Barrios, 2016).Labour scarcity has also been reported to inhibit uptake of ingenious agricultural technologies (Black et al., 2011).Deane et al. (2013), on the contrary, present the argument that migration paves the way to technology innovation and development in the countryside through the unearthing of new ideas and knowledge and investment of accumulated capital repatriated by returnee labour migrants.Supplementary research reports also point out to the finding that households of labour migrants are more inclined to adopt new farming skills, methods and processes than those of non-migrants on increasing agricultural productivity (Mendola, 2008;Ratha et al., 2016).Labour migration has been contemplated by households to augment earnings and support the livelihoods of members of households living in rural areas (Semyonov and Gorodzeisky, 2008).Resource accumulation is the most crucial element in migration and rural development (Ratha et al., 2011;Mendola, 2010).
The increase of studies on migration and rural development in migrant-sending areas has received newest interest by policy makers since early 2000 (de Haas, 2010;Adams and Cuecuecha, 2013).A common agreement has been circulating around which endorses the argument that return migration together with earnings accumulated and invested in rural areas of origin contribute substantially to reducing poverty and boosting the standard of living of the people in the countryside.Households of migrants (particularly those with capital invested back home) have proved to have earnings and expenditures of an upper level opposite to those of nonmigrants (Schmook and Radel, 2008;Wouterse and Taylor, 2008).
In spite of the significant potential of return migration for rural development, limited research has been conducted to understand the determinants and impact of return migration.The few available studies have not yet been able to generate important information on returnee labour migrants and reintegration as well as effective utilization of remittances.Additionally, the topic of return migration between Tanzania and Europe is particularly under researched, with only a limited number of empirical studies on return migration to Tanzania.The need for gathering sufficient and quality data on return migration in Tanzania cannot be ignored if a satisfactory analysis of this topic is to be achieved.The requirement of carrying out a diligent and systematic research into this subject matter is highly important and timely.This paper attempts to close this gap by contributing to the existing literature as part of a wider debate on migration and rural development in Tanzania.To this end, the objective of this paper is to determine whether or not return migration can influence rural development in Tanzania.The specific objectives of this study were to: (i) determine the probability of returnees becoming entrepreneurs.(ii) assess the extent to which re-migrants employ themselves in the agricultural sector and (iii) assess the extent of involvement of re-migrants in local development projects.
The paper sought to advance the scholarly discussion on human labour mobility and rural development by methodically examining three separate hypotheses.The first hypothesis states that native labour migrant returnees from distant urban centres have a high probability of being involved in entrepreneurship in contrast to migrants whose purpose for moving out is not getting employed.This dichotomy of conception is anticipated owing to the fact that precedent studies indicate that native labour migrants are anticipated to demonstrate higher levels of self-employment versus migrants who move out for purposes other than employment.The reason behind this contention is that, as local inhabitants, they stand a better position to utilize the local circumstantial elements known to affect selfemployment with certainty (Naudé et al., 2017).
In examining closely the extent to which return migrants are likely to engage themselves in agriculture, a second hypothesis was formulated which posits that native labour re-migrants from well endowed rural areas closer to their homelands are expected to employ themselves in the agricultural sector contrary to taking non-farm selfemployment.To some extent, this anticipation is attributed to the fact that short distance migration is presumed to be cheaper contrary to long-distance migration (Ghatak et al., 2008;Msigwa and Mbongo, 2013).Considering such disparities in expenses incurred, groups with a low stock of skills and knowledge from rural areas are thus deemed to make use of short-distance migration to well endowed and benign rural areas for all intent and purposes of accumulating capital in support of agricultural self-employment back home.The third hypothesis states that native migrant returnees are more likely to involve themselves intensely in volunteer local development projects belonging to the societies of their rural native lands.This anticipation is grounded on the fact that migrant returnees have been found to engage themselves in community developmental projects like school renovations and construction of rural health facilities (Chukwuedozie and Onokala, 2013).

THEORETICAL FRAMEWORK
There is an assortment of theoretical paradigms for contemplating return migration.Each of these presents a different postulation which mainly comes from theories some of which are neoclassical economics, structural approach, trans-nationalism, the new economics of labour migration and social network.This study focused exclusively on one of the economic theories namely "The Structural Paradigm to Re-migration".This relational model presents that return is not only examined in relation to the observation that the individual"s experience of the migrant is of prime importance, but also the analysis is in relation to social and institutional factors in heritage countries.This paradigm underlines the influence of the financial and economic resources repatriated back to the homeland following the decision to go back in addition to the reunification of the re-migrants to their local societies.However, this structural framework does not only conceptualize return as an individual"s experience of the migrant but contends that return migration should also be examined within the social and institutional context in the sending country (de Haas, 2010).
Cerase in Kunuroglu et al. ( 2016) provides four contrasting kinds of re-migrants, paying attention to their ambitions and expectations."Return of innovations" undoubtedly forms the most active classification of remigrants in Cerase"s classification.This category refers to migrants who are "prepared to make use of all the means and new skills they have acquired during their migratory experiences" in the hope of attaining their expectations in their places of origin which, in their understanding, offer lucrative opportunities to meet their aspirations.Cerase comments that these re-migrants consider themselves to be ingenious; for they tend to take it that the skills obtained overseas in addition to their saved earnings will have turned them into "change agents".The advocates of this theory (Hunter, 2010;Farrell et al., 2012 ) present that returnees" success or failure is analysed by establishing a relationship between the condition of the people and the economic status of the home and the aspirations of the re-migrants.In the structural approach, Gibson and McKenzie (2009) contend that the resolution to re-migrate is to a great extent tied to factors such as family, belonging to an ethnic community, and home attachment, contrary to the income opportunities available in overseas countries.

Study area
The territory covered for the study comprises four rural villages drawn from Nzega and Magu districts (Figure 1).The villages taken from Nzega District include Itobo and Mbogwe while those taken from Magu District include Nsola and Ngasamo.They show uniform environmental and agriculturally suited climatic characteristics.Nzega District is located between latitudes 4° 20' 00" south of the equator and longitude 33° 05' 00" East of the Greenwich meridian whilst Magu District lies between 2° 30' 00" South of the equator and longitude 33° 30' 00" East of the Greenwich meridian.The population inhabiting the rural villages used in this study according to the 2012 population census is made up of 298,171 males and 295051 females.
The study communities were picked commensurate with two criteria.The first one considers the high magnitude of rural-to-rural, rural-to-urban and across national border labour mobility; and secondly the distance from the district headquarters as a criterion for the picking of the rural areas.Both of these criteria escalate the integration of rural labour out-migration and the rural environment and thus can assist in the understanding of how rural urban labour migration influences rural development.

Sample size and sampling procedure
The multi-stage sampling procedure was applied in this study.In the first stage, a list comprising all households was compiled in the company of leaders of villages.Nsola had 387; Ngasamo, 813; Itobo, 358 and Nata, 530 households respectively.The total households were 2088.The second stage involved subdividing the households into two categories subject to their labour migration status.Households having a minimum of one migrant member were categorised as labour-migrant households, whilst those consisting of all relatives present for the period were assigned as non-labourmigrant households.The total number of households with labourmigrants in all four villages was 832.According to the 2012 Population and Housing Census of Tanzania (URT, 2013) Nsola village had an average household size of 5.5, Ngasamo had 5.8, Itobo had 4.9 and Nata had 5.1.Assuming the average sizes of these households did not change significantly over the 6 year period since the last census was taken, the population who are potential candidates for sampling would be 847 for Nsola, 1879 for Ngasamo, 701 for Itobo and 1076 for Nata village; thus making a total of 4503 for the 832 households with labour-migrants.Each labour-migrant household was required to produce one migrant who was eligible for the interview.
For valid inferences, the minimum sample size was calculated using a formula by Godden (2004) which helped to arrive at a representative number of respondents when the population estimate is known. (1) Where; n = sample size z = z -value a (e.g., 1.96 for a 95% confidence level; p = the proportion of occurrence of the variable of focus (which is 0.5 where the figure is not known b ); c = marginal error or confidence interval of ±4 % (±0.04).
A Z-values (Cumulative Normal Probability Table) stand for the probability that a sample will fall within a certain distribution.
The Z-values for confidence levels are: 1.645 = 90% confidence level; 1.96 = 95% confidence level; 2.576 = 99% confidence level.Substituting for the formula; n = (1.96) 2 .(0.5). (0.5).1/ (0.04) 2 = 600.25.Since the target population was below 50,000 the following formula was used; (2) Now the population being 4503, therefore, the new sample size = 600.25 but considering a refusal rate of 8%, the final estimated sample was rounded up to 648 respondents all drawn from the labour-migrant households.The proportionate allocation of eligible respondents was 122 for Nsola, 270 for Ngasamo, 101 for Itobo and 155 for Nata.Before commencing the study, the institutional ethical clearance was observed and taken.

Data sources and design
This study employed descriptive cross-sectional design adopting quantitative and qualitative approaches for data collection, synthesis, examination, and elucidation.During the process of collecting the primary data for this research, the household questionnaire was administered to the respondents so as to derive information on households" demographic conditions, migrants" attributes, uses of savings accumulated and saved in destination areas by the return migrants, returnees" entrepreneurial investments and rural societal development schemes accomplished by re-migrants in their rural places of origin.After the completion of the survey, the filled questionnaires were checked for errors, completeness, and accuracy before embarking on the processing and analyzing exercise.
Descriptive statistics were applied for analysing the socio-economic and demographic profiles of re-migrants.A multivariate analysis technique mainly the probit model was used to assess the probability of entrepreneurship and engagement in rural community development projects with the help of STATA 9.In addition, one key informant and one opinion leader well-thought-out to be adequately knowledgeable in each of the sampled rural villages were interviewed.Consequently, two interviews were administered Masanja 77 in each study area giving a total of four for both districts.Interviews were administered to gather ethnographic information, especially those which may be cumbersome to be adequately captured by questionnaires because African communal inclinations do not often find a suitable representation numerically (Nzeadibe et al., 2012).As a result, illustrative quotes from key informant interviews (KIIs) were used to reinforce the quantitative data.

Analytic model and empirical specification
The unit of analysis or subject of the study is a re-migrant who is 18 years and above.The reasonable ground for interviewing individuals with an age of 18 years or older lies on the fact that in Tanzania, the age of consent is 18 years and is considered that people of this age and above are physically and mentally mature.Return migrants mean those people who worked overseas or domestically at another richer area at least for one year or more and returned to this country, more specifically to their areas of origin within the past five years.The coming back to their native home places may be either temporary or permanent.
Entrepreneurship is governed by many factors such that the set of written questions in the questionnaire was split into three segments to facilitate understanding of the re-migrants" circumstances ahead of out-migration.An additional inquiry was made on knowledge, and skills garnered overseas and undertakings in the native home community after coming back from overseas.Entrepreneurship relies chiefly on the re-migrants" demographic attributes, education, skills, experiences, new values, ideas and savings accumulated overseas.Subsequently, in the questionnaire, the first sub-division gathered data that included re-migrants" age, sex, birth-place (classified by rurality or urbanity), work sector in the rural area of origin, education level, reasons for migration, and total migration expenses incurred.
The second section collected information on total overseas savings, skills acquired overseas and period of stay in the foreign country.The third sub-division solicited information concerning usage of saved earnings which was categorized into 4 sections namely home consumption (food, clothing, house construction), social services consumption (health, education), entrepreneurial consumption (business investment, commercial agriculture), and volunteer local community development schemes (school renovations and construction of rural health facilities).
Return migrants" probability to become entrepreneurs was gauged by applying the Probit model.In this particular analysis, an enterprise is regarded to be a dependent or response variable whilst re-migrants" demographic characteristics, earnings garnered overseas, the duration of stay and skills acquired overseas are captured as independent or predictor variables (Gubert and Nordan, 2011;Wahba and Zenou, 2012).
However, this study added some predictor variables such as education level, marital status, types of destination countries and length of stay abroad and time since the migrant returned to the area of origin.These points are important for explaining entrepreneurship.This paper used these variables to see what determines entrepreneurship and engagement in community development projects among return migrants.
Analytically, entrepreneurship differences were examined among migrants.Statistics on individual"s involvement in entrepreneurship were then applied to construct a dichotomous dependent variable (entrepreneur = 1; otherwise = 0).This dependent variable was utilised in the Probit regression model which is specified below.

𝑛 =
In this model, Y is denoted as the response variable measuring the outcome of becoming an entrepreneur after coming back, and ß0… ß8 are the independent variables or explanatory variables and e is the residual.
Age, on the other hand, expresses return migrant"s age, Gen is a short form for gender which refers to masculinity or femininity, which again is a dual variable with assigned values of male = (1) and female = (0).Dstay means duration or period of stay overseas or in another intra-country area which once again is a continuous variable.In the model, Skill is considered as a latent variable expressing the degree to which a re-migrant acquired new competences overseas.ReturnT refers to the length of time since migrants returned to places of origin and ReturnC indicates from which country or domestic area a migrant returned.ε is the error term (Devkota, 2016).
The model applied carries three basic assumptions.First, returnees who accumulated sufficient financial resources overseas are more inclined to become enterprisers upon their return.Likewise, Migrants who tend to live abroad for a long-term can save more money.Saving stimulates investment.Second, educational qualification also influences entrepreneurship.If a returnee is more educated, he/she is more likely to allocate more saved money for investment and less for other consumption headings.Third, fresh returnees are unconvincingly unable to turn entrepreneurs contrary to those who came back home over a year ago.It is also expected that a bigger household size is better positioned to offer a migrant than a nuclear family.Finally, skills obtained overseas are also essential for enterprising in the area of origin.This study also applied the qualitative method to investigate the importance of a return migrant"s role in development projects such as construction and renovations of societal facilities in their own localities.

Salient socio-economic characteristics of the respondents
The social, economic and demographic profile of the respondents is anchored on variables such as age, gender, education status, marital status, family size and whether returned to his/her rural area of origin.Salient attributes of the respondents are shown in Table 1.
The first column in the table presents a list of predictor variables.Subsequent columns present values for the mean and corresponding standard deviation.Survey data indicate that re-migrants" age ranged from 25 to 62, whereas the mean age is 36.Of the total number of remigrants, only six percent were females whereas 94% were males.The majority of the returnees (63%) had moderate family sizes, followed by small families (27%) and large families (10%).
In excess of half of the surveyed migrants (56%) had not finished secondary level education, whereas the remainder went over and above secondary education.Average overseas saved earnings in the past 5 years were higher (four times) than the average non-migrant savings in Tanzania.The mean period of time lived by out-migrants overseas was about 5 years.Majority of return migrants (52%) got employed in the construction (machine operators, painters, electricians) manufacturing and domestic (home, restaurant, and hotel) industries.A noteworthy proportion (72%) of re-migrants acquired new skills.Returnees from the US, Germany, Poland and the UK accounted for 43% while internal migrants (57%) came from mining areas.Sixty-seven percent of these returnees returned to their places of origin one year before the interview, while the remaining proportion returned 3 to 5 years earlier to the survey.

Estimation of results
The first hypothesis which states that migrant natives returning from distant urban centres have a higher probability of being involved in entrepreneurship contrasted to non-migrants was tested utilising variables such as the overall savings, the span of stay, the percentage of total savings repatriated and the skills learned overseas.Probit results (Table 2) indicate that all these variables showed a positive marginal effect, further implying that these factors could trigger the starting of commercial activities and as a result re-migrants with such qualities stand a better chance to becoming entrepreneurs.A study by Rivera and Reyes (2011) established that migrants" overseas money transfers back home were unfavourable (but significant) in motivating entrepreneurship.The repatriation of accumulated savings back to their hometowns, actually, could beregarded to be statistically significant at 10 per cent level.This consideration strongly suggests that the longer the period of time overseas migrants take to work overseas, the higher the chances that they will start up an enterprise in their home country.Table 2 further presents the Probit estimates of remigrants turning into enterprisers.Results show that overseas earnings saved, and individuals who returned back home earlier than 2 to 5 years ago in Tanzania in addition to those who returned back home during the time preceding the same span of time from intra-country mining areas are extremely important for entrepreneurship at the level of one percent.Amongst remigrants, those who stood a good chance of being entrepreneurs were youthful males.Female migrants on the other hand were likely to emerge entrepreneurs after return, holding other conditions unchanged.
Turning to the variable of age, its anticipated influence on entrepreneurial behaviour is unclear.Age was found to be unfavourably correlated with the probability of self-Y = ß0Age + ß1Gen +ß2Mar + ß3Hsize, ß4Ed + ß5S + ß6Dstay + ß7Skill + ß8ReturnT + ß9ReturnC + ε employment and in particular entrepreneurship.This result was found as a contradiction to Marchetta (2012), who argues that older people due to their larger professional experience and holding more connections have a higher probability of turning an entreprenueur.With regard to marital status, results show that married individuals had more chances being entrepreneurs in comparison with single groups.This finding is similar to what Luchyk (2017) concluded in his study of Polish returnees, that is being married reduces the chances of non-participation in the labour market and increases the chances of being entrepreneurs after the return.The probability of changing into an entrepreneur after returning to the home of origin seems to be exceedingly higher in both categories of returnees such as those who got a career and technical education during the time they were overseas and for those who came back home without the influence, control or guidance of others.People who received more education are apparently likely to become entrepreneurs compared to re-migrants bearing less education.First, the well-educated migrants stand a better chance of getting employed in high wage paying jobs in the host countries which translates to additional income resulting in added savings.Secondly, they keep updating themselves on ongoing development schemes at home as they communicate with close friends or significant others from time to time.Communication performs a fundamental function for entrepreneurship.Thirdly, people that are well-educated have the managerial capacity of a higher quality than people with little education.This study finding is in agreement with the results of several previous studies (Gubert and Nordman, 2011;Hamdouch and Wahba, 2012).The aggregate earning saved overseas stands out to be another notable element for the likeliness of entrepreneurship occurrence.More investment practice has been observed amongst the old re-migrants (those who came back home over two years ago) than fresh returnees.In contrast to theoretical predictions, sending remittances back home was not found to be a factor responsible for influencing the probability of a migrant returning.However, what the study found was that the stronger the family ties a migrant had with the home of origin, the more likely the migrant is to return.Similar evidence was found by Hedberg (2009); Bijwaard, Schluter, and Wahba, 2012;and Pungas et al. (2012).Some migrants having lived overseas for a long time, have succeeded collecting money, and eventually returned to their homes of origin with some business plans.An anticipated favourable outcome of being an enterpriser before migration is clearly built for returnees: Previous enterprisers are about 29 per cent destined to continue being enterprisers after returning to their homes of origin and this result supports the findings of Piracha and Vadean (2010); Démurger and Xu (2011); Giulietti et al. (2013).Furthermore, the key informant interview in Nsola Village confirms the findings as one returnee from the U.S.A. stated: I successfully started an irrigation farm as a commercial enterprise and have reached a point where I am now a reliable supplier of vegetables in the horticultural industry.Additionally, I have managed to distribute fruit and plant seedlings to over 2000 families in the village with an aim of supporting small-scale farmers to engage in fruit irrigation farming thus increasing their incomes and livelihoods together with presenting growth opportunities for them.Source: Key informant interview, CM.
The second hypothesis posits that native migrant returnees from richest natural resource-laden rural areas which are closer to their home villages have a higher chance of being engaged in agricultural self-employment in contrast to the rural non-farm economic sector.This hypothesis was tested using two variables, namely savings accumulated from mineral-rich rural areas and the probability of purchasing more land for agricultural undertakings accommodating irrigation.Results show that return migrants from the well-endowed rural areas, especially mining areas are characterized by certainty at the level of 10%; it is suggestive of a significant effect that warrants investing in agriculture.This result does not support the finding by Luchyk (2017) who reported that males from the rural areas in general had a smaller chance to become self-employed in comparison to individuals from cities, even though working in the agriculture is associated with the highest chances of being agriculturally self-employed.Production in mining areas is generally irregular and consequently hard to determine.Nevertheless, by looking at the amount of wealth that has been collected and the proportions of disbursement by the possessors of pits (some maintain to be paying out to the tune of Tshs.2,000,000/=($887) per month to support human resources and operate the equipment), their earnings are quite substantial.The average earnings per month per mineshaft was estimated to be Tshs.2,000,000/=($887).This earning indicates that the area of variation between upper and lower limits is substantially higher in the mining areas contrasted to the non-mining areas.Subsequently, higher earnings result in higher investment possibilities.All things being equal, the results of marginal effects indicate that a oneunit addition in the aggregate number of internal remigrants from rich rural areas who have accumulated monetary capital increases the probability of land purchase by 20% (Table 3).Lastly, the value of Pseudo R-Squared stipulates that independent variables in the Probit model predicted 34 per cent of lack of consistency in entrepreneurship.This result corroborates the findings by Davis and Lopez-Carr (2014) and Devkota (2016).
Table 4 shows the estimates of hypothesis three which tests returnee"s engagement in rural local development schemes in home villages.It was tested using variables, for example, the duration of stay overseas, the percent of total savings repatriated home from abroad and the skills learned abroad.Results show that the intensity of rural development investment appeared to be independent of age, size of land and also the value of productive assets.
In analysing the results, it was observed that the sociodemographic variables of the respondents such as gender and marital status have no influence on the remigrants" participation in community development schemes.The findings of this study echo previous qualitative researches which argue that when re-migrants come back to their countries of origin, they bring new and innovative thoughts with them, and consequently, promote social development to their communities.Tobit results of this study (Table 4) corroborate the findings of Black and Costaldo (2009) who by using data from Ghana and Côte d"Ivoire, found that return migrants are more likely to start a new enterprise if they had accumulated savings and stayed abroad longer.The results imply that returnees may also accomplish this undertaking through non-monetary means.One example is by eye-opening local members of the communities to alternative ways of looking at the world.
The resultant quantitative trends derived from this study parallel Waddell"s (2013) on-the-ground research in Guanajuato (Mexico).As an illustration, while interviewing re-migrants in the field, this study found they were expected to promote the importance of education.In Itobo which is located in Nzega District, Tanzania, remigrant leaders demonstrated a feeling of pleasure by showing the school that they had volunteered to fund and renovate.One key informant from Mbogwe village (a returnee from Poland) also narrated how he teamed up with members of his local community to construct a feeder road that would help ferry farmers" produce to a major road.He said: I was born in this village; I felt it was necessary for all of us in this village to have access to the trunk road so that we could take our produce to different market places.That's why I have spent part of my savings accumulated abroad to hire a bulldozer in the construction of this road.I'm happy that my fellow villagers have offered their physical strength to cooperate with me in this project.Source: Key informant interview, SS.
After separating out the effect of socio-demographic characteristics together with human abilities and skills variables, return migration was found to be statistically significant in ascertaining entrepreneurship, self-farm employment, and community development project engagements.While the data analysis of this study appears to have supported hypotheses1, 2 and 3, some exceptions need to be reiterated.These include cases of overseas money transfers back home which were found to be unfavourable since the response of the current policy in Tanzania has been slow in harnessing return migrants" potential to create positive impacts and rural development.Likewise, sending remittances back home was not found to influence the probability of a migrant returning.Considering the empirical evidence of this study, the success rates of these covariates as noted by other similar studies cast doubt as none of these are robust factors and consequently further research along these lines is necessary.

CONCLUSION AND POLICY IMPLICATIONS
This study has demonstrated that re-migration has the capacity to augment community-driven improvement in socio-economic conditions and prospects.Nevertheless, as earlier researches have disclosed (Arce et al., 2011), in most situations modern skills that earlier migrants return with are underexploited.Fundamentally, the extent to which re-migrants affect their native land may, for the most part, be dependent on local people who are involved in making policies.The current low potential use of a re-migrant"s resources is likely to be a consequence of local elements such as the absence of state endorsement, inadequate bond markets, and poor infrastructure.Given these premises, the findings from this research have salient future effects for people involved in making policies.As a means of making the best use of development returns of the collective set of skills, knowledge, and experiences of a human being brought back by re-migrants, it would be most beneficial to the Tanzanians to encourage returnees and decrease possible impediments to development within regions with high migration flows.
The results of this study have indicated that in areas where re-migrants settle down in life, non-migrants get the chance of inclusion in development initiatives.Likewise, health facilities get renovated, additional classrooms are constructed and rural roads are initiated.These research findings help expose the capacity of remigrants in contributing to various development schemes in their home area communities.These novel findings contribute to the growing body of knowledge and literature related to the impact of re-migrants on communities of their native homelands.Local governments in Tanzania therefore can design local economic development policies which are suited to inviting departing and returnee migrants to invest in their villages and also in updating them with on-going government funded or proposed development undertakings in local communities.The local governments can as well initiate entrepreneurial policy reforms and special support geared at creating new enterprises in the localities.Finally, local legislation of new enterprises coupled with land tenure reforms can altogether empower local governments to promote and register a viable environment for business and investment in the remigrants" home areas.

Figure 1 .
Figure 1.Map of Tanzania showing regions and location of study areas by districts as indicated by arrows.

Table 1 .
Salient characteristics of return migrants.
Source: Author"s calculation based on Return Migration survey data (2017).

Table 2 .
Probit model of becoming an entrepreneur after return.

Table 3 .
Marginal effects for the probability of investing in agriculture after return.

Table 4 .
Probit estimates of engaging in rural development projects.