Home  /  Blogs


Indian labor regulations take small steps in the right direction

Devashish Mitra's picture
There have recently been some important changes to India’s labor regulations. These are small steps in the right direction as the country tries to reduce the number of informal workers. These changes include Rajasthan’s recent amendment that raises the threshold for seeking state government permission for retrenching or laying off workers from 100 to 300 workers. Rajasthan has also increased the threshold employment for stipulations on work hours, work days and minimum age requirements under the Factories Act. And they have raised the minimum membership for the registration of a representative union from 15% of the firm’s employment to 30%.

In addition, in a movement away from the Inspector Raj, the central government has set up a unified web portal where firms can themselves file their compliance reports pertaining to 16 central labor acts.  Such a web portal is expected to have a built-in algorithm that will select firms that need to be inspected. Hopefully, this is just the beginning and we will see more reforms in the future.

But why is this so important for India? Developing countries tend to have a large proportion of their employment in the informal sector. In India it is especially high. For instance, the informal percentage of employment in non-agricultural employment is 83.6%. The proportion of informal employment in the manufacturing sector is 87% (ILO, 2012).  If you use per capita GDP as a predictor and compare India with other countries, India’s predicted informal employment as a percentage of non-agricultural employment should be a little above 60% (ILO, 2012). This means that, given its level of development, the actual proportion of informal employment in India is about 40% higher than it should be. For India to reach that level of 60%, the formal sector would need to more than double in size.

Another way to look at this is by comparing correlations between poverty and informal employment. About 28% of the Indian population is below the national poverty line. Brazil, Uruguay and Costa Rica have poverty rates close to that number. However, their share of informal employment in nonagricultural employment is around 40%. In Uganda, where the poverty rate is also roughly 28%, the share of informal employment stands at around 70%.

India-China comparisons are always interesting – even though China’s per capita income is much higher and its poverty rate much lower. Here we can see that China’s informal non-agricultural employment share is only 32.6%.

Why is India’s informal sector so large? Here it is extremely important to note that most of India’s labor regulations apply only to the formal sector and to formal workers. The large informal sector is most likely a response to these regulations.

In the World Bank (2006) Doing Business survey, India and Mexico are the only two countries that answer in the affirmative to six of the seven questions on the existence of rules governing redundancy dismissals and layoffs. In addition, only India, Indonesia, Mexico and Sri Lanka require third party approval for dismissals. Thus, India’s labor regulations are probably among the most restrictive in the world.

There are 200 labor laws in India, including 52 Central Acts. Probably, the three most restrictive acts are the Industrial Disputes Act (IDA), the Industrial Employment (Standing Orders) Act and the Trade Union Act. The IDA requires firms with more than 100 workers to seek permission from their respective state governments for retrenchment or laying off of workers. This is seldom granted. The Industrial Employment (Standing Orders) Act requires such firms to ask for permission even for modifications in job descriptions. The Trade Union Act lets any seven workers within a firm form a union, which leads to multiple labor unions. In addition, this act provides each such union the right to strike and to represent workers in legal disputes with employers.

It is important to look at the data on the impact these laws have on overall employment, the average quality of jobs and exports of labor-intensive manufactures. Several empirical studies have looked at variation in labor laws and their implementation across Indian states, which have arisen from state-level amendments. The following are some important findings regarding the impact of labor regulations.

  • Besley and Burgess (2004) find that state-level formal manufacturing output, investment, employment and productivity in India declined with restrictive or rigid labor regulations. The corresponding aggregates for informal manufacturing increased.
  • Hasan, Gupta and Kumar (2009) find that Indian states with relatively restrictive labor regulations have experienced slower growth of labor-intensive industries and overall employment than others.
  • Dougherty, Frisancho Robles and Krishna (2014) show that productivity in Indian firms in labor-intensive industries as well as in industries facing highly volatile demand was, on average, about 11%-14% higher in the states with less restrictive labor laws than in others.
  • Hasan, Mitra and Sundaram (2013) find that India uses more capital-intensive techniques of production than predicted by its level of development and than those used by China in many industries including paper and printing, leather, rubber and plastics, chemicals, non-metallic minerals, base metals, metal products, electrical equipment and instruments and petroleum.

Thus we see that India’s outdated labor regulations have restricted formal nonagricultural employment. The recent initial steps towards reforming labor laws, though in the right direction, form just a tiny fraction of what needs to be done. The blocking of further progress in labor reforms by recent labor union protests is, therefore, somewhat concerning.

Informality and formality – two ends of the employment continuum

Shanthi Nataraj's picture
Informal employment predominates in many low-income countries and is here to stay, possibly for a very long time. Such employment includes casual and temporary workers, day laborers and domestic workers. It is also the owners, unpaid family employees and workers in household enterprises and other small firms that fly under the government’s radar. Although many developing countries have experienced strong economic growth in recent years, their share of informal workers has remained stubbornly high.

Informal employment may even increase in coming years. Many youths in developing countries continue to be informally employed, and there is growing informalization of formal sector jobs due to contract employment, casual labor and other arrangements. Traditional expectations for a stable, full-time job as the norm are also being challenged in the developed world.

The apparent staying power of informality is a concern because informal employees tend to be paid lower wages, and to receive fewer benefits and legal protections than formal employees. Thus, a critical issue for policymakers is how best to improve conditions for those who remain informally employed.

One potential option is to extend formal labor regulations to a wider set of firms that would include informal employers. But enforcing these regulations could be prohibitively costly for low-income countries. It may also be counterproductive: a number of studies from low-income countries show that making it more expensive for firms to hire employees formally – for example, by requiring termination notices or higher wages – can reduce formal employment. This is particularly true for vulnerable groups such as youths, unskilled workers and women. It also can end up increasing employment outside the regulated sector.

Therefore, it becomes critical to recognize that informality and formality are two ends of a continuum. It is also critical to identify those specific aspects of formality that workers value most, that employers would be most willing to provide, and that governments would find most feasible to enforce. In fact, many workers classified as formal by standard criteria do not have access to the full suite of benefits associated with formality, while many informal workers do have one or more such benefits.

In Bangladesh, formal employees are more likely to receive many types of benefits, but benefit receipt is by no means an all-or-nothing phenomenon. For example, 20% of formal workers do not have access to sick leave or paid holidays, while about half of casual workers and 10% to 20% of day laborers do. About half of formal workers report receiving termination notice, as do about 20% of casual workers.

Source: Author’s calculations based on Bangladesh Labour Force Survey 2009-2010.

These are the dimensions of formality that we can, and do, measure. Perhaps even more critical are the dimensions that typically remain unmeasured such as adequate workplace safety, freedom from coercion and access to legal recourse in the case of mistreatment; formality may also help protect workers from mistreatment or abuse. When a firm or its workers are registered with the government, there is a greater chance that basic health and safety regulations may be enforced, that conditions such as child or bonded labor may be hard to conceal, and that workers may be able to appeal to authorities when faced with unfair labor practices.

An important issue for policymakers to consider is which dimensions of formality to target for expansion. Which aspects do workers value most about a formal job? Is it a contract, termination notice, paid sick leave or time off? Is it the level of pay, stable hours, or the right to appeal to authority in case of poor working conditions or contract violations? Which benefits would employers be most willing (or find least costly) to provide, and might be least likely to push more employment into the shadows? Would certain benefits be easier for governments to encourage and monitor – for example, through random spot checks in industrial clusters – without requiring small firms to register and thus become subject to the full suite of formal regulations?

Developing a better understanding of the aspects of formality that are most valuable to workers, are easiest for small firms to adopt, and are most likely to be enforced is a critical step toward designing more targeted and effective policies to improve working conditions for the majority of workers in low-income countries.


Education, jobs and economic development in Bangladesh

Wahiduddin Mahmud's picture
Bangladesh is about to benefit from a demographic dividend because of a youth bulge. The challenge for the education system is to leverage the rapid growth in the labour force into higher economic growth performance. This youth bulge, combined with the successful campaigns for universal primary education, is leading to huge increases in the supply of semi-educated labour. There is enormous potential for utilising this workforce productively by expanding post-primary education and training on the one hand, and by creating commensurate employment opportunities on the other hand.

Bangladesh’s economic growth so far has been driven by a replication approach in respect of low-productivity ready-made garment export, export of low-skilled labour, and expansion of micro-enterprises.  For the next stage of growth, we need to switch from replication to innovation. We need to increase productivity and develop skills. For this, the education system needs to be geared towards developing well-balanced people with appropriate skills and flexibility for adjustment. This will let them keep pace with increasingly competitive and globalized markets and rapidly changing technologies.

Educating broadly or deeply?
Less developed countries facing resource constraints face a choice between educating their population broadly or deeply; either they can adopt a universalist approach of providing basic education for all or an elitist approach of emphasising quality higher education for the most talented. In East Asia, a universal approach to education in part led to high economic growth driven by manufacturing exports and without worsening of income distribution.  On the other hand, India’s ability to take advantage of the new possibilities in high-tech information services largely resulted from its long-standing investments in higher education.

Now, however, less developed countries like Bangladesh may not have to make this choice between the two alternative approaches to education. China and Vietnam, for example, have expanded primary and higher education simultaneously, recognising that success required both universal literacy and a cadre of highly skilled individuals capable of absorbing advanced technology.

The prevailing higher education system in Bangladesh has led to educated unemployment. There is an apparent paradox here. Not only has higher education been increasing rapidly, but there also seems to be excess demand for such education. The evidence of excess demand can be seen from several factors: the increasing and higher economic returns to education at the tertiary level relative to lower levels; the rapid proliferation of private higher education; and an increasing number of students opting to study abroad. Yet, the unemployment rates remain unacceptably high among educated and skilled workers.

The educational planners thus have to address a number of questions: Is expansion of higher education leading only to more educated unemployment because of absence of appropriate signalling by the labour markets? What kinds of skills are in shortage? What reforms in the education systems are needed to make college graduates employable, and to address the skill shortages? Clearly, the prevailing systems of higher education do not seem to incorporate enough technical and employability skills that could directly lead to better labour market outcomes.

The skill mismatch in the labour market is also related to a country’s capacity to take advantage of opportunities in the global markets, such as through technology adoption and development of new export industries.  Many technologies imported by the less developed countries from more advanced countries may not find suitable local workers, causing labour mismatches. Adapting these technologies to local conditions requires even more skills. Policies and institutions need to be in place to mitigate this mismatch and to make sure that the imported technologies are eventually adapted to local conditions.

There are also important equity issues to be addressed. Increasing returns to higher education along with unequal access may lead to a deepening of income and social inequalities. Studies have shown that the degree of access to education has replaced family ownership of land and other assets as the main vehicle of transmission of poverty and inequality from one generation to the next. The Bangladeshi education system has not been able to respond adequately to increasing demand, let alone extend access on an equitable basis. How far children from poor households can compete in a merit-based system of entry into higher education will depend on their access to quality education at the primary and secondary levels. So far the policy focus has been on getting these children to school in the first place. The time may have come to shift emphasis from the global agenda of education for all, to providing access to quality education and to higher education for the children from disadvantaged families.



Creating employment in post-conflict and fragile states

Nisha Arunatilake's picture
This blog looks at the job creation in the fishery sector in the Eastern Province of Sri Lanka, after the ending of the armed conflict in this province in 2007. The Eastern Province is a resource rich province. It has huge areas of inland water.

​Traditionally, fishing has been a main economic activity given its 420 kilometers of coastline. Moreover, 5.5% of the province is covered in irrigation tanks and canals. In 2013, 21% of the total population of about 1.5 million in the province were living in fishing households.

The Eastern Province is ethnically the most mixed province in the country, with Tamils, Muslims and Sinhalese comprising 40%, 38% and 22% of the population respectively.  The armed secessionist conflict left Eastern provinces in economic disarray. Most of the economic infrastructure in this region was destroyed or damaged, and the traditional livelihoods of individuals were disrupted.  In 2007, around 4% of the population was displaced due to the conflict and a further 2% was displaced due to the tsunami that took place in December 2004, Department of Census and Statistics data show.

The ending of the war in 2007 in the Eastern Province, paved the way for development. But, the conflict had left the region vulnerable and fragile, making employment generation a challenge. A primary barrier for employment was security. Intermittent attacks by LTTE rebels made it necessary for the government to keep a large security force in the East (International Crisis Group, 2008).  The tensions between the Tamils and the Muslims over security and land continued (International Crisis Group, 2008). Despite the ending of the armed conflict, the heavy presence of the forces, periodic disappearances of people and politically motivated killings left the community insecure. Also restrictions on livelihood activities such as fishing, cattle herding and gathering of firewood were impediments to economic development.

With the ending of the conflict the government launched a three-year rehabilitation and resettlement programme named the Eastern Revival (Nagenahira Navodaya or NN). The main objective of this programme in the first 180 days was to make lands accessible through de-mining, reconstructing damaged infrastructure, resettling the displaced and helping them restart their livelihoods. The second phase concentrated on developing the region by improving economic infrastructure including, electricity, roads, transport, water and sanitation.  A major part of the NN programme was aimed at creating jobs and livelihoods in the province. The proposed plans included the development of the Trincomalee harbor, through a public-private partnership, and the fishery harbor at Oluvil in the Ampara district.

The most challenging aspect of resettlement was clearing mines and providing security to the resettled. NN was able to resettle all internally displaced persons, in keeping with international standards, by the end of 2009.  Rehabilitating the damaged infrastructure was also essential for the free movement of people. Projects under the NN improved connectivity between towns and communities via the construction and rehabilitation of more than 3000 kilometers of roads and 30 bridges (Central Bank of Sri Lanka, 2007).

Statistics from the Ministry of Fisheries and Aquatic Resources Development show that the infrastructure available for fishers has increased since 2007. For example, the area under perennial tanks used for inland fishing has increased from 30,155 Ha in 2011 to 37,446 Ha in 2013. This increase accounts for 45% of the total increase in the area under perennial tanks in the country during the 2011 to 2013 period. Further, the production capacity of active ice plants in Kalmunai, Batticaloa and Trincomalee in the Eastern province increased from 160 Mt/ day to 420 Mt/day from 2008 to 2013. This accounts for 44% of the increase in capacity of active ice plants in the whole country.

The above developments were able to increase the fishing jobs in the province from 66,600 in 2008 to 89,530 in 2013. These account for roughly 40% of the increase in fishing jobs in the country during the same period (see Graph).

Source: Own construction using Ministry of Fisheries and Aquatic Resources Development data.
Note:  Districts marked with an asterisk are from the Eastern province.

The creation of employment in the fishery sector was influenced by many factors. But its success mainly depended on the simultaneous attention given to clearing lands, resettling families, improving security, and provision of economic infrastructure. Without attention to all these factors, the success in improving employment in the fishery sector in the Eastern province would have been more gradual.


Jobs and conflict: solutions from Andhra Pradesh

Sanjay Pulipaka's picture
It is well known that successful counter-insurgency operations need to have a development component. However, the challenge is getting the right mix of policy measures. An examination of experiences from India may help us identify appropriate and effective counter- insurgency strategies. Many states in India are experiencing Naxalite insurgency. Drawing inspiration from Communist ideology, particularly Maoist ideology, the Naxalites believe that an armed insurrection to overthrow the state is an important step to achieve a classless society.

In 2013, the Government of India declared that there are 106 districts which are experiencing Naxal related violence. Today, the most impacted states are Chhattisgarh, Madhya Pradesh, Orissa, Bihar and West Bengal. Since 1980, more than 15,000 people lost their lives due to Naxalite related violence. Close to half of these deaths (6,700 deaths) happened during 2005-2015 and this year 130 people were killed in Naxalite related incidents.

While various states in India have been struggling to contain Naxalism, there was a sharp decline in the fatalities caused by this left-wing Extremism in Telangana and Andhra Pradesh. In 2005, the total fatalities were 320 and this came down to just four in 2015. What explains the decline of Naxalism in the Telangana and Andhra Pradesh regions in India and what are the lessons to be learnt? Specifically what was the interface between job creation and decline of Naxalism?

The state government adopted a three-pronged strategy in response to Naxal related violence. First, the government of Andhra Pradesh did not depend on the police forces from the Union government. Instead, it created various specialized units such as the Greyhounds to carry out counter-insurgency operations. By increasing recruitment, the police force was also strengthened in terms of numbers as well. For instance, the total strength of the Andhra Pradesh police force went from 63,662 in 1993 to 107,730 in 2013.

Second, the state government sought to create employment opportunities at the grass roots. For instance, there were 299,484 school teachers in 1997, which increased to 495,478 in 2011. Further, various schemes of the central government such as the Mahatma Gandhi National Rural Employment Guarantee Act also added impetus to the state government’s effort to create employment at the grass roots. In 2013-2014, under this scheme 6,565,827 households were allocated work in Andhra Pradesh and of this, nine districts in Telangana regions accounted for approximately 2,880,773 households. Further, all the districts of Telangana (except Hyderabad) received additional funding under Backward Grant Regions Grant Fund (BRGF) of Rs10 crores ($1.5 million) per district. A substantial portion of these funds were earmarked for local infrastructure projects such as rural water supply, construction of social welfare hostels and rural electrification.

The government of Andhra Pradesh also initiated construction of large numbers of irrigation/hydro-electric projects (30 major and 18 medium irrigation projects). Simultaneously, a massive rural housing program called the Integrated Novel Development in Rural Areas and Model Municipal Areas was launched in 2006. This resulted in the construction of almost 617,769 house units in Phase I. While there have been allegations of corruption, the projects generated significant employment opportunities in the short-term.

Third, Hyderabad, the capital city of Andhra Pradesh, emerged as a prominent software hub. The number of employees in Information Technology sector in Hyderabad witnessed an exponential increase from about 8,700 people in 1997-1998 to approximately 341,268 in 2012-2013. All these new jobs might not have gone to the people from Andhra Pradesh. However, it should be noted that according to the study conducted by NASSCOM, a job in the IT sector creates about four jobs in other sectors. The emergence of the IT sector and the spillover impact in other sectors generated a strong perception in the state that the private sector employment opportunities are on the rise.

The Andhra Pradesh experience is broadly in consonance with Paul Collier’s argument that construction of infrastructure gives high returns in facilitating a movement from conflict to post-conflict situations. Further, the question is not if private sector can generate sustainable employment opportunities in conflict situations but where should the private sector be deployed in conflict situations. In areas that are under government control, the private sector should be promoted. The growth of private sector led employment, even in small geographic areas will add momentum to politics of aspiration as opposed to politics of grievance. And in areas that are under the shadow of the insurgency, government-led employment schemes will yield better results.

Education and Employment: The big push needed for India’s youth

Raghbendra jha's picture
When looking at recent data, it is hard to escape the conclusion that although India has enjoyed high economic growth this has largely been jobless economic growth. It is imperative for there to be a big push in the areas of education and employment in India. This is the most significant policy challenge facing the Indian economy.In 2012, India’s population was 1.23 billion of which 65% was of working age.  India already has a smaller ratio of old people to those of working age population than that of China. Therefore, over time, if India’s youth is productively engaged, the country’s private financial savings and physical capital investment are likely to boom.  In contrast to China, India’s population will continue to grow beyond 2025 and these trends are likely to persist well into the future. By this time, India can be a high-income or high middle-income country.  Arguably, no country currently faces such fortuitous circumstances; indeed very few countries ever have.

Mass education of youth and their gainful employment in productive jobs is central for capitalizing on India’s demographic dividend. But India’s performance with regard to both education and employment has been disappointing. Except for the duration of compulsory education and hours of instruction for pupils aged 9, India’s performance in primary education is lacklustre.

Data from WDI2013 reveal that in 1999 only 63% of male students and 60% of female students who had begun grade 1 reached grade 5. This is a lower rate than for lower middle-income countries. The WDI data also indicate low rates of labour participation by youth (particularly women) and high rates of unpaid family work.

New policy framework

Much of the new job creation must come from areas other than the farm sector. Average farm yields in India are low. In 2012, they were on average 2.3 tonnes per hectare largely because of the small size of farms.  A nuanced policy of creating 115 million non-farm jobs along with large scale investment in agricultural infrastructure could help raise yields to 4 tonnes per hectare by 2022.

Most of the new jobs will need to be created in states that have poor initial educational and infrastructural conditions.  This will require large increases in public and private investment, a supportive reform program and be centered on manufacturing and construction.  Here are some suggestions:

  • Design an integrated approach to India’s infrastructure/construction needs as one of the key initiatives of Niti Ayog – the organization that has replaced the Planning Commission.
  • The plethora of labor and product market regulations (for large and small businesses) must be reduced. They inhibit labor mobility and adaptation to domestic and global market requirements and, more broadly, increase the administrative cost of doing business.
  • One stop clearances for projects and a strong culture of e-governance need to be cultivated.
  • Tax distortions must be reduced and a well harmonized Goods and Services Tax introduced.
  • Other product market distortions that impede cross-state flow of goods and services must also be reduced.

A critical input into such job creation would be the rapid skilling of India’s youth. Apart from a substantial revamping of school level education, this would require sharply redesigned and expanded national apprenticeship programs.  While a large part of these would be established through regular classroom contact, there would be a significant role for remote learning through, for instance, the newly formulated Digital India initiative.

The investment needs for such large scale job creation cannot be fulfilled by the private sector alone.  A program of fiscal prudence would lower the fiscal deficit and – at the margin – switch public expenditure from current subsidies to investment. This would facilitate private investment and FDI so that the investment rate can be edged closer to 40%. This is the rate of investment that is required to sustain high rates of growth and employment creation.  The savings rate must commensurately climb to ensure that there is no instability from a large current account deficit.

Perhaps the most significant change required among policymakers is attitudinal – both in the public and private sectors.  The current fixation with growth and poverty is understandable. However, India’s development philosophy must realize that that neither high, medium-term growth, nor sustained poverty reduction, are possible without a paradigm change in India’s approach to the education and employment of youth.  The consequences of failure could be grave.

Professor Raghbendra Jha, is Head of the Arndt-Corden Department of Economics, at the College of Asia and the Pacific, of the Australian National University in Canberra. 


If Sri Lanka is to join the knowledge economy, it needs to improve its education, training and skills

Nisha Arunatilake's picture
With innovation taking a central role in driving markets, countries are increasingly looking to invest in innovation and technological change to be competitive and improve productivity. Innovation is driven by talent and creativity. But the demand for highly skilled workers, especially workers in the science and technology fields are increasing globally. 
Sri Lanka too is aspiring to join the bandwagon of innovation. The Mahinda Chinthana – Vision for New Sri Lanka announced in 2010 – gives emphasis to knowledge-based economic activities, which require skilled workers. Skilled workers are needed not only to develop new technologies; they are also needed to identify uses for existing technologies and to make use of them in work places. The stock of qualified skilled workers in the country is limited. Many workers who carry out the work of skilled workers are not formally qualified for the jobs that they perform. A recent IPS study shows that only 15% of individuals working in specialized science and technology field jobs have a degree (see figure 1). 

Source: IPS calculations based on Labor Force Survey data 2009
Increasing the stock of skilled workers in the country is challenging for several reasons. First Sri Lanka’s workforce is ageing.  In 1990 only 35% of workers were aged 40 or above, but by 2012 this proportion has increased to 52%. This has serious implications for the long-established pattern of schooling, training and work.  Even today, most training programs in the country are geared towards school leavers and the young. Only a handful of training opportunities are available for mature workers. With fewer young workers entering the labor force and the older workers receiving limited formal training the stock of skilled workers are not expanding.
Second, high-end training opportunities are limited in the country. Only a very small proportion of the young people who leave school enter university (see figure 2).  A similar small proportion of individuals receive vocational training. Only 17% of youth in the 20 to 24 age group are participating in education. This proportion drops to less than 3% for 25 to 34 year olds and to less than 1% for 35 to 44 year olds.
Third, insufficient numbers of high school graduates have a background in science and technology related subjects so that they can enter science and technology-related training programmes.  Labor shortages in certain industries have driven the government to expand training opportunities in some sectors. But the demand for such training programmes is limited, as many do not have the basic general education pre-requisites.  Part of the problem is in the lack of science-educated high school graduates. This again stems from several shortcomings in the system because access to science education is limited in the country. Also the low quality of education at junior schools prevents children from obtaining the necessary grades in mathematics and science at younger ages to qualify for science education at higher grades.
Figure 2: Education participation by age group

Source: Own calculations based on Labor Force Survey 2009
Fourth, many qualified science and technology workers emigrate. This is partly due to aggressive recruitment campaigns carried out by developed countries to attract them. Many young who temporarily migrate to foreign countries to study, end up migrating permanently. Poor salaries for skilled workers and lack of challenging job opportunities are main push factor for emigration for skilled workers. Many aspiring parents also opt to live abroad to give better access to good quality schools and universities for their children.
If Sri Lanka is serious about moving towards a knowledge-based economy several things need to happen simultaneously. Quality of education needs to improve from kindergarten to post-graduate level. For if attention is only given to the university sector, there will not be sufficient numbers of qualified youth for the universities.  The education sector needs to give special attention for promoting science education and harnessing creativity. Last of all, investments that create more lucrative employment opportunities need to be promoted to keep talented workers in the country. 

Nisha Arunatilake is a Research Fellow and Head of Labour, Employment and Human Resources Development research at the Institute of Policy Studies of Sri Lanka.

Business training for small enterprises in developing countries: does it work?

Suresh de Mel's picture

What helps small businesses grow? In January 2009, two colleagues (David McKenzie and Chris Woodruff) and I began an experiment that examined the effects of general business training on current and potential female small business owners in urban Sri Lanka. For international comparison purposes, we focused on the most commonly used general business training programme – the ILO’s Start and Improve Your Business (SIYB) programme. This training programme has been given to over 4.5 million people in more than 100 countries in more than 40 languages.

How the study was structured

  • We focused the experiment on two samples of females aged 25-45. This was because a previous experiment which examined the rates of return to capital injections into micro-enterprises, also conducted in Sri Lanka, had shown that female entrepreneurs tended to lag behind their male counterparts.
  • The first sample was a group of 628 women who worked more than 20 hours per week in self-employment, in a sector other than seasonal agriculture or fisheries, and earned monthly profits less than SLR 5000 (around US$ 40).
  • The second sample was a group of 628 women who were out of the labor force, but were planning to enter self-employment in the next year. They were able to identify the type of proposed business as an indication of their seriousness of intent.

We refer to these two samples as current and potential business owners respectively. The baseline survey was conducted in January 2009 (Round 1).

From each of the two samples, we randomly chose 400 respondents to be offered business training. The potential business owners received the Generate Your Business (GYB) and Start Your Business (SYB) modules over 8 days. The current business owners received a refresher GYB and Improve Your Business (IYB) modules over 6 days. At the end, both groups joined together for an additional day of technical training to provide exposure to some relatively high-return business sectors which are also socially acceptable for women in Sri Lanka to engage in.  These included sectors such as food manufacturing, beauty culture, sewing clothes and plant nursery.  In each group, half of those who completed the training were also given a SLR 15,000 (around US$ 130) grant.  The training programmes were conducted in April/May 2009 and cash grants were distributed in June 2009. The remaining 228 respondents in each sample formed the control group.

Follow-up Surveys
Four rounds of follow-up surveys were conducted in September 2009 (Round 2), January 2010 (Round 3), September 2010 (Round 4) and June 2011 (Round 5). This corresponds to 3-4, 7-8, 15-16 and 24-25 months after the training. Attrition rates were relatively low, ranging from 6%-12% over a period of 2 ½ years.


  • Training leads to improvements in business practices among current firms. We used a composite business practices score to reflect management practices related to marketing, stock control, record-keeping and financial planning. We found that training leads to improvements in these business practices in both the short term (Round 2) and medium term (Round 4 and 5). These improvements come in all the four components of the business practices score.  But the improvements in business practices are quite modest. The training only and training plus cash groups were only implementing an additional 2 or 3 business practices respectively (out of a maximum score of 29).  Current business owners are still a long way from implementing the practices taught in the SIYB programme.  Business practices are the management practices related to the four areas of marketing (0 to 7), stock control (-1 to 2), record-keeping (0 to 8) and financial planning (0 to 12).  This gives an overall score in the range of -1 to 29.   Business Outcomes refer to four measures of firm performance – monthly profits,  monthly sales, capital stock and hours worked in the business by owner.
  • Training alone has no impact upon business outcomes. We examined the effect of training on four business outcomes – profits, sales and capital stock of the business, and hours worked by the owner in the business—and found no significant impacts. These results held up across all follow-up rounds so this cannot be attributed to a timing related issue either.
  • Training and cash grants together have a positive impact on some business outcomes. The impact of training plus cash grants upon capital stock was positive and significant and held across survey rounds. In terms of magnitude, the increase in capital stock was close to the SLR 15,000 grant received. Profits also show a significant positive increase in the short term (Rounds 2 and 3). But in the medium term (Rounds 4 and 5) the magnitude is much smaller and loses significance. This suggests that the effect on profits is only temporary.
  • Both treatments speed up entry into self-employment among the potential owners.  After receiving training plus cash grants, potential business owners were 29% and 20% more likely to be running a business by Round 2 and Round 3 respectively. Among those receiving training only, the likelihood of operating a business by Round 2 increased by 12%. However, the impact of both treatments fell and lost statistical significance by Rounds 4 and 5. The positive impact on self-employment rates among potential owners is short-lived and disappears in the medium-term.
  • The treatments affect who is drawn into self-employment among the potential owners. Training drew less analytically skilled women (as measured by a Raven test) into operating a business. Training plus cash grants drew in women with less analytical ability and less household wealth.
  • Training has an impact upon business outcomes of potential owners.  The training resulted in a 40% increase in sales and 43% increase in profits in the medium term. Short term impacts are not significant.


  • Generating re-investable profit growth in subsistence-level female-run businesses remains a challenging task; the constraints seem to lie beyond the training and capital spheres.
  • In contrast, training does seem to help potential owners to open businesses more quickly and helps to improve their business outcomes. This has implications for women’s livelihood development.
  • Our study also demonstrated the importance of tracing progress over time; relying on a single follow-up survey would have led to very different conclusions.

Dr. Suresh de Mel is a Senior Lecturer in the Department of Economics and Statistics at the University of Peradeniya in Sri Lanka.

Fixing India’s labour law cholesterol

Manish Sabharwal's picture

Changing the ways that India’s benefits regimes are funded would be the most impactful way to reform the country’s labour market. It is not changing the law to make it easier for companies to hire and fire employees.When people think of labour law reform in India they tend to think of the laws around hiring and firing; specifically, reforming Chapter 5B of the Industrial Disputes Act.​But there are many other reforms possible:​​

  • collapsing the 44 central labour laws into five clusters;
  • fixing our dysfunctional benefits regime;
  • making trade unions more representative.​
  • making compliance for employers frictionless;
I would also argue that the most impactful labour law reform would be fixing our benefits regime. Under this regime, the mandatory deductions to gross wages are 45% for an employee with a Rs5500 monthly salary but only 5% for an employee on a Rs55,000 monthly salary. Making matters worse, this 45% deduction from low wages goes to programmes that give poor value for money; 55 million of 100 million Employees’ Provident Fund of India (EPFO) accounts are dormant. Additionally, EPFO’s administrative costs of 440 basis point make are one of the most expensive in the world for a government securities mutual fund.

Even more painfully, most of the employer contribution to EPFO now goes to the bankrupt Employee Pension Scheme (EPS) whose $8billion hole is being balanced by reducing benefits. Furthermore, the Employees’ State Insurance (ESI) has India’s worst health insurance claims ratio, paying only 49% of contributions as benefits. It is also sitting on $5billion of idle financial investments.

I would propose the following reforms, which would create three choices for employees in how their salary is paid:

  • ​Choice 1; Paying or Not Paying the 12% EPFO employee contribution
    The 12% employee contribution is an unaffordable salary deferment by low wage employees who have no savings. Employees should be allowed to opt out of this contribution at joining, pay this into their individual National Pension Scheme (NPS) or continue the status quo by paying to EPFO.
  • Choice 2; Paying the 12% employer contribution to EPFO or NPS       
    Currently most of the employer contribution to EPFO goes to the Employee Pension Scheme (EPS). Employees must be allowed to choose between EPS or diverting their entire employer contribution to their individual National Pension scheme individual account.
  • Choice 3; Paying health premiums to ESI or an Insurance company 
    Employees should have the option to pay their monthly health insurance premium to ESI or buy a policy from any IRDA-regulated health insurance company.

It is well known that India’s labour laws have negative consequences. The country suffers from poor productivity, poor working, conditions lower taxes and a large number of small, informal firms. Indeed manufacturing comprises just 12% of employment and 85% of manufacturing comes from firms with less than 50 employees.

Nevertheless, the country has finally begun to fix this regulatory cholesterol that has been holding back formal job creation. In the last year the new government has already amended the apprentices act, revamped the labour inspector regime, given employers and employees unique numbers, and is moving to online compliance. Most importantly the central government now allows state governments to diverge their labour laws from the national norm. This means Rajasthan’s reform lead is now being followed by seven other states.

But changing the rules of the benefit system would be the most impactful reform that the government could now undertake.

Manish Sabharwal is Chairman, Teamlease Services, based In Bangalore

The manufacturing conundrum

Aashish Mehta's picture

Jesus Felipe [1] is an Advisor in the economics and research department at the Asian Development Bank.
Aashish Mehta [2] is an Associate Professor of Global and International studies at UC-Santa Barbara.
Changyong Rhee [3] is a Director in the Asia and Pacific Department at the International Monetary Fund.

In a recently released working paper [4], we have analyzed data on manufacturing employment and output levels from developing and developed countries.  We show that historical manufacturing employment levels are much better predictors of subsequent prosperity than historical manufacturing output levels. Unfortunately, we also find that late industrializers are likely to face significant difficulty achieving high manufacturing employment shares, even if they are successful in spurring high levels of manufacturing activity. Therefore, while manufacturing jobs are key, it will be difficult for today’s late industrializers to replicate past success with development through manufacturing job creation.

A long tradition in development economics treats manufacturing as special.  Some of the development benefits the sector provides rely on large and growing numbers of manufacturing jobs, not just rising levels of manufacturing output. Labor productivity in manufacturing converges rapidly towards international norms. This gives aggregate productivity a bigger boost; the larger the share of the workforce that is in manufacturing, the greater the boost. The more rapid the manufacturing employment growth, the faster workers move out of low-productivity, low-wage agriculture and traditional services.

Unfortunately, cross-country analysis of the connections between manufacturing employment and levels of development has been restricted to OECD countries. This matters because many developing country governments have large programs to stimulate manufacturing activity, on the understanding that jobs and higher incomes will follow.  Ambitious job targets are announced – such as 100 million new manufacturing jobs by 2022 in India. These are typically justified with reference to the experiences of earlier industrializers, like Korea and Taiwan.  Public budgets, land and labor regulations, and even education policy are being modified to pursue these manufacturing jobs.  We can see why developing countries want these manufacturing jobs: our data show that a country’s peak manufacturing employment share between 1970 and 2010 rather than is its peak manufacturing output share, is a much better predictor of its average per capita GDP in 2005-2010.  Controlling for peak manufacturing employment shares and the date that manufacturing activity peaked, peak output shares are insignificant predictors of subsequent prosperity.  This suggests that manufacturing output matters for prosperity only insofar as it comes with jobs. Moreover, we show that every country that is rich today, by any reasonable standard, had more than an 18-20% manufacturing employment share sometime since 1970.

It is important to note that industrial activity typically grows with income in poorer countries, peaks, and then falls with income and wages in richer (deindustrializing) countries.  There are two reasons we think that manufacturing employment-led development is becoming more challenging.

  1. Labor productivity has risen faster in manufacturing than in the wider economy. Higher levels of manufacturing output are now compatible with lower levels of manufacturing employment. Figure 1 confirms this, showing that peak manufacturing employment shares have fallen over time. Peak output shares have not.Figure 1: Manufacturing Employment Shares Have Declined, Output Shares Have Not
    Figure 1: Manufacturing Employment Shares Have Declined, Output Shares Have Not
  2. Manufacturing activity is now more apt to leave for other countries as labor costs rise. Therefore deindustrialization kicks in at lower income levels. Moreover, this premature deindustrialization is more apparent in employment than in output data. Output can be sustained in the face of rising labor costs by replacing workers with machinery. (Arvind Subramaniam and Amrit Amirapu show similar trends [5] in industrial (manufacturing plus mining, utilities and construction) employment using repeated cross-sections of countries.)

Countries still industrialize and then deindustrialize as they become richer.  However, industrial employment shares for today’s late industrializers such as China, India and Bangladesh are all below 16%, and on today’s trends seem unlikely to rise much further. Moreover, the per capita income levels at which deindustrialization kicks in have fallen from $34,000 in 1970 to around $9,000 in 2010.

These results urge a balanced approach to industrialization. They confirm that industrialization matters – when it brings jobs; but they also confirm that this is less and less likely to happen. Governments must not neglect manufacturing. Nor can they rely as heavily on it as they once did.



Tackling social exclusion in the labor market

Rebecca Holmes's picture

Rebecca Holmes is a Research Fellow in the Social Protection Program at the ODI

The general trend in poverty reduction over the last decade for many countries in South Asia has been a positive one. However, look a little closer and for particular groups of the poor there has been little opportunity to move out of poverty. In many contexts, there is a strong correlation between poverty and being excluded from the economy and society because of gender, caste, ethnicity or religion.

Crossing a foot bridge. Photo: Shehzad Noorani / World BankSocio-economic norms fuel gender discrimination, where women are less mobile and therefore unable to migrate to find work. They are also responsible for domestic chores limiting economic opportunities, they receive lower pay than men and are less likely to own assets and less likely to participate in community decision-making. Take for example, women in the Chars – a remote area in northern Bangladesh. Facing multiple dimensions of exclusion on the basis of their gender and geographical remoteness, these women live in an area which has been historically politically and economically marginalized, resulting in few economic opportunities and highly seasonal work.

In our recent ODI research [1] we examined the extent to which social protection and labor market programs can tackle social exclusion in south Asia. Specifically, we examined the asset transfer program called the Chars Livelihoods Program [2] (CLP). We can draw three important lessons from the program, which tell us how best to support poor, socially excluded households in the labor market.

The first lesson is perhaps the simplest. Interventions must be designed and implemented appropriately to respond to specific needs. A clear example of this can be seen in the CLP example in Bangladesh. In recognition of both the sociocultural and the economic barriers that women face in generating income and accessing and owning productive assets, CLP transfers a large-value asset – such as a cow. This is culturally appropriate for women to generate an income from as it can be done from their homestead.

The second lesson is that access to complementary services and programs is necessary. Ensuring labor market participation for the socially excluded requires more than just income opportunities. The CLP program in Bangladesh demonstrates that an integrated approach has been key to its success. This approach not only supports women through direct asset transfers, but it also provides training opportunities to strengthen their skills and knowledge, as well as providing opportunities for creating social networks both within the community and more broadly. Indeed, our research shows that CLP has had a positive impact on women’s livelihood diversification and ability to generate an income from agricultural-related activities. However, we also find that there are limitations to what one program can achieve. Tackling wider issues of exclusion and marginalization is necessary to enable greater (and more equal) labor market participation. Focusing on improving women’s skills alone is insufficient to enable them to take advantage of economic opportunities.

Another key lesson to emerge from our research is the importance of supporting the empowerment, agency and voice of the socially excluded. A strong body of literature [3]exists which demonstrates the correlation between empowerment and economic productivity. While the programs across our case studies in south Asia sought to improve the income generating opportunities for marginalized women, none of them explicitly sought to challenge the imbalance of power structures which cause and perpetuate exclusion and discrimination in the labor market. No significant changes were evident in women’s decision-making power, or making demands to program or government officials. However, there were some positive indications that indirectly programs can build women’s confidence in interacting with community members or local government officials as a result of increased income or contact with program implementers or the government through program channels (see below).

Changes in respondent’s self-perceived confidence levels in interacting with community members and officials after receiving CLP benefits
(% of beneficiaries)

Changes in respondent’s self-perceived confidence levels in interacting with community members and officials after receiving CLP benefits

Source: ODI




Literacy Unplugged, Bollywood-style

July 02, 2014 | Contribution by BRIJ KOTHARI 

Brij Kothari is Professor of Communication at the Indian Institute of Management, Ahmedabad, and the Founder of PlanetRead.

Source:  From JKP : []

In India, the formal sector accounts for only 6% of the total labor force of nearly 500 million. This means that the bulk of labor participation is in the informal sector, trapped in a vicious cycle of low skills, low wages, and low productivity. If these individuals are to have a fair chance at not only upping their job skills, perhaps through vocational training, but also leveraging these gains on social and emerging technological fronts to escape their cycle of poverty, they must have basic functional literacy (not just nominal “literacy”). However, the vast majority of these individuals do not. The encouraging news is that there are numerous efforts under way to dramatically turn this situation around — including a successful program of using subtitles for Bollywood movies (see “Better Late than Never” in Education and Skills 2.0: New Targets and Innovative Approaches, 2014).

Photo credit: Still from Jodhaa Akbar with Same Language Subtitling, Copyright Disney UTV. 
Note: Subtitle in image reads “In the folds of these moments.”

How Literate is Literate?

Although independent India’s literacy rate rose from 18% in the first national census (1951) to 74% in the most recent round (2011), an astonishing 50-60% of the so-called “literates” cannot read the day’s newspaper headline, or a Grade 2 level text, in their own language. That is not to say that they are fully illiterate either, because most of them can identify at least a few letters and are, therefore, best thought of as, “early-literate” but functionally illiterate. How high is functional illiteracy? Based on my research (with Tathagata Bandyopadhyay; see Can India’s ‘literates’ read?, 2011) and ASER’s findings year after year about the poor state of reading achievement in Grade 5, I estimate that around 400 million “literate” Indians are actually functionally illiterate. This is in addition to the 273 million who are officially illiterate.

India’s “ability to read a newspaper headline rate,” if there were such a measure, would hover in the 30-37% range. No government would like its galloping literacy rate to be punctured with a statement like this. But if a government does not acknowledge the reality of its nation’s literacy quality, it is unlikely to do much about it. A big problem is that the literacy rate in many countries, like India, is simply measured by asking individuals to self-report for all household members if they are “literate” or “illiterate.” Even a slight variation on the question, like “Can you read a newspaper?” would give an accurate measure of functional literacy. We have found that people’s response to this question is highly correlated to tests that measure an ability to read any simple text functionally. People report accurately because the question is specific and, in their perception, easily subject to verification. One could even replace “newspaper” with “bus board” or “letter” as proxies for functional literacy.

While governments do focus on the challenge of getting illiterates off the starting block, they are swift to label this achievement as “literacy” because all the incentives revolve around getting the literacy rate up. From the government’s perspective, the functional literacy rate is best left unmeasured lest the problem surface officially. The state, therefore, gets off the hook of having to also plan for the more arduous and longer transition of early-literates to functional literacy.

Improving Literacy, Bollywood-style

How does one transition 400 million early-literates in India to functional literacy? The essence of a solution lies in creating conditions that allow for at least a few minutes of easy reading engagement, every day and throughout life. This is possible if reading itself becomes an integral part of something people do every day, like, watching television. In India, 750 million people watch, on average, two hours of TV every day. Bollywood-style films and film-based content is a dominant genre in a large number of Indian languages. What if we subtitled the lyrics of all existing film songs on TV — in the same language? Word for word, what you hear is what you read. This was a question we first posed in 1996, calling our approach “Same Language Subtitling” (SLS). It causes automatic and inescapable reading engagement among early-readers (and readers alike) whenever they happen to watch film songs with SLS. Anyone with some letter familiarity cannot but try to read along, as confirmed by a body of eye-tracking research. Viewers like to read along to songs for its Karaoke-like experience and to know the song lyrics.

Since 2006, SLS has been implemented on 10 weekly half-hour song-based TV programs, in as many languages, currently delivering reading practice to 200 million weak-reading TV viewers in India. As we have slowly scaled up, we have done rigorous testing to ensure that SLS is working, and we now have strong evidence that it is a proven solution on the scale that is required. The Indian Institute of Management, Ahmedabad (IIM-A) and the Nielsen-ORG Center for Social Research studied the effects of SLS in 3,179 households during 2002-2007, demonstrating that exposure to 30 minutes of SLS per week increased the functional literacy rate from 25% to 56% among students with at least five years of schooling (see PlanetRead for the latest studies). Plus, the cost is minuscule. For example, with a viewership of 20 million, one U.S. dollar can give 30 minutes of daily reading practice for about 1,000 people per year.

IIM-A and PlanetRead are now on course to deliver regular reading practice to all the 750 million TV viewers in India at present, and growing rapidly. More than half the viewers are also weak-readers. The goal is to implement SLS on all songs on TV in India, in all languages, through national policy. The strategy is to scale up in India first and let that speak as a model for expansion to other countries, especially in South Asia and Sub-Saharan Africa, on popularly watched song-based programming, in the local language. The latest update in this narrative is that the Broadcasting Corporation of India (Prasar Bharati), the national TV network (Doordarshan), and the Planning Commission have supported, in principle, the scaling up of SLS nationally.

In India and many other countries, the problem of low quality of literacy is real, if under-researched, and seldom acknowledged officially. Yet we feel strongly that the SLS solution is proven and cost-effective. So what is holding back broadcast policy in India and other low literacy nations from considering SLS seriously? What is holding back private networks with a genuine interest in doing “well” (ratings do go up) by doing “good?” These questions are not easily answered. But at its core, policy-making generally lacks the risk-taking ability required to advance social innovation in a time-bound manner. On the other hand, private networks may need to do much more to bring — what might sometimes be a peripheral interest in doing good — to a place where doing well and doing good are inseparable.

A range of high level experts will offer their views on issues of key job challenges by writing blogs. The suggested categories for blogs are: