Linda Calabrese (Senior Research Officer, ODI) & Stephen Gelb (Principal Research Fellow – Team Leader, private sector development, ODI)
27 June 2017
Recently The New York Times published an article by Christopher Blattman (Columbia University) and Stefan Dercon (Oxford University and DFID) questioning the poverty-reduction effect of sweatshop work in developing countries. They carried out a randomised experiment, which ended in 2013, with almost 1,000 Ethiopian jobseekers, placing some in factory jobs in one of five factories, providing a second group with some entrepreneurship training and a modest grant, and leaving the rest to find an income however they could. One year down the line, two thirds of the first group had left their manufacturing jobs, while those who remained were working longer hours, earning less and facing more work-related health hazards compared with those in the second group.
Blattman and Dercon conclude that their study shows factory work is not an ‘escalator out of poverty’, saying ‘everything we believed [before the study] would turn out to be wrong’. They argue that the study has shaped their views of factory work: ‘In the short run, workers seem to share few of the benefits but a heavy burden of the risks’ from industry. This is hardly surprising for those who know the industrial sector in developing countries today (remember the Rana Plaza fire in Bangladesh 2013?) or indeed the history of manufacturing in now-industrialised countries (remember Dickens?).
Apart from the well-known methodological problems of Randomised Control Trials (RCTs), as experimental studies are known, we do not think that one study of 1,000 people and five factories in one country can or does tell us nearly enough about the costs and benefits of industrialisation as a development path. And as Blattman and Dercon themselves acknowledge, ‘we simply do not know of any alternative to industrialisation. The sooner that happens, the sooner the world will end extreme poverty.’ They suggest that the difficulties faced by the factory workers in their study resulted from deficiencies in the businesses – bad, or at least very inexperienced, managers – and the absence of policies providing social protection.
Blattman and Dercon’s findings are echoed in our recently published paper on foreign direct investment in Myanmar from China and elsewhere, which examined the clothing and shoe industries amongst others. Myanmar’s income per capita is around double that of Ethiopia. We did not do any experiments, but we conducted several dozen interviews of firm managers, local representatives of clothes buyers such as large European retailers, and NGOs working with management and workers to upgrade business performance.
The clothing industry has expanded rapidly since EU sanctions on imports from Myanmar were lifted in 2012, and further expansion is expected because US sanctions were lifted last September. The industry employs close to 200,000 workers in about 340 firms, of which about 180 are foreign-owned, though many that are officially locally owned have silent or hidden foreign partners.
One of the attractions of Myanmar for the garment industry is that wages are low while productivity is relatively high – not as high as in China, but above most African countries, according to the managers we spoke to. Managers complained to us that high worker turnover was one of their biggest challenges, along with unreliable electricity, bad roads and the difficulty of finding skilled workers. A study conducted on a sample of less than 200 firms showed that in one year, the average firm lost around 40% of its workers – with peaks of 57% in the garment sector. Many workers live in slums outside the industrial areas, with no access to water or electricity. Rural–urban migration flows push many rural dwellers into urban centres.
There are factories with unpleasant working conditions for the predominately female and young workforce – very hot factories, often poor or no drinking water or sanitation facilities, long working hours. But the story is mixed – we also visited a Hong Kong-owned factory supplying garments to the UK and Europe, where workers enjoyed a good working environment including a canteen and on-site medical staff. Since it opened, this factory has experienced very low turnover.
Learning how to manage
Blattman and Dercon point out that their own intervention in the hiring process of the five factories they studied introduced a degree of organisation unknown to the managers. This again is unsurprising – new industries, new factories, so also new managers. In Myanmar, we found the same: foreign factories employ foreign managers almost exclusively, and domestic managerial skills are lacking. Interestingly, in one Asian-owned factory we visited, the managers and technicians were mostly from Madagascar and Mauritius, African countries that have developed successful garment industries and are now exporting their skills.
Scarce management capabilities are undoubtedly one of the major constraints facing development. We recommend in our report that Myanmar prioritise developing technical and managerial skills by setting up tertiary training institutes specifically for the garment industry, as was done in Bangladesh. Foreign managers provide a short-term solution, but in the long run Myanmar and Ethiopia need to develop their own talent pools.
The path to better wages and working conditions: pressure from below and from above
Myanmar also shows how the industrialisation path itself can lead to improved wages and working conditions, as political dynamics play out within the garment sector. On the one hand, the large workforce enables organisation and collective action, which presses governments to regulate labour markets better and raise standards. Trade unions were made legal in Myanmar in 2012, and union pressure and workers’ strikes contributed to the introduction of a minimum wage in 2015. At 3,600 kyat (less than $3/£2) per day, the minimum wage level is among the lowest in the region, which is significant for firms’ international competitiveness. Before the introduction of the minimum wage, many – probably most – factories paid their workers much less, and the low wages forced them to work long hours to top up meagre incomes with overtime.
Pressure from workers’ organisation – ‘below’ – is complemented by pressure from ‘above’. The garment sector globally is dominated by large retailers and clothing brands, and many of these buy clothes produced in Myanmar: H&M, Primark, Marks & Spencer and The Gap, for example. In fact, we found that these large buying corporations are often significant in influencing existing suppliers in China or elsewhere in Asia to start up a production operation in Myanmar, so they are contributing to the country’s industrialisation. And the buyers are also very important in influencing supplier factories’ behaviour. The buyers face consumer (and NGO) pressure from their customers in rich countries, who do not want goods made by exploited, unsafe or insecure workers or produced by child labour. Retailers and brands demand in turn that their suppliers maintain good labour standards – and they have the power and influence to monitor and enforce such standards. For example, global retailers supported the introduction of the minimum wage.
Little surprise, then, that a large systematic survey of garment firms (Tanaka, 2017) showed that employment and safety conditions, wages, union recognition, fire safety and health care were better in exporting firms (almost all with official or hidden foreign ownership) than in non-exporters. Turnover rates were also lower in exporting factories, possibly because they offer better labour conditions.
Of course, not all firms are exporters, but exporters show the way by creating an upward pull that, together with worker demands, places pressure on non-exporters, at least those in the same industry. Eventually, as current circumstances in China illustrate, and as Blattman and Dercon acknowledge, these upward pressures force firms to adopt new strategies – to introduce new technology with improved productivity and higher incomes, or indeed to shift to lower-wage locations and start the cycle of improvement there.
Blattman and Dercon do not discuss the broader industrial and political context in which ‘their’ five factories operate in Ethiopia. But we would surmise that Ethiopia is at an even earlier stage on the path than Myanmar, so that some of the dynamics already strongly in play in Myanmar are only just emerging in Ethiopia.
Entrepreneurs out of necessity, or factory workers?
Of course, we recognise that this upward path is not inevitably followed, and neither is progress along it smooth and linear. There are setbacks for workers even in rich countries, as we see with the recent spread of the ‘gig economy’ and zero-hours contracts. And the industrialisation path excludes many people, at least from its direct benefits: the 200,000 employed in Myanmar garments is a large number but not nearly enough to absorb the 56% still working in agriculture or the millions doing informal work in the cities.
Such ‘necessity entrepreneurship’ by people aiming to survive is – realistically – the only option for many, even most, people in developing countries, and will remain so for a long time yet. Blattman and Dercon found that the people in their study who were given a short business training and a small grant had a slightly higher average income at the end of the one-year study than those in the factories. This is positive, but is it enough for a ‘solution’, enough for either poverty reduction or sustained income growth? Industrialisation drives strong growth in incomes and productivity, and we would argue that these benefits are essential to lift informal incomes as well. The need for social protection systems that are adequate in both their levels and their coverage is crucial, as Blattman and Dercon insist. But, to reiterate their conclusion, the sooner industrialisation happens, the better for ending extreme poverty. It is not a choice: the response to the challenges of industrialisation is not to forego it but to do it faster and better.
This blog has been released alongside a briefing and longer study on foreign direct investment and economic transformation in Myanmar which can be found here.
 Bernhardt, T., Kanay De, S. and Thida, M.W. (2017) Myanmar labour issues from the perspective of enterprises: Findings from a survey of food processing and garment manufacturing enterprises, ILO, CESD, GIZ.
 Theuws, M. et al. (2017) The Myanmar Dilemma: Can the garment industry deliver decent jobs for workers in Myanmar? SOMO, ALR, LRDP.
 See, for example, Qualitative Social and Economic Monitoring (2016) A country on the move: domestic migration in two regions of Myanmar. World Bank Group.
 Zajak, S. (2017). ‘Trade union building in Myanmar’, Open Democracy, 17 February (https://www.opendemocracy.net/sabrina-zajak/trade-union-building-in-myanmar).
 Reuters (2015) ‘Myanmar sets $2.8 daily minimum wage in bid to boost investment’ (http://www.reuters.com/article/us-myanmar-economy-wages-idUSKCN0QY0A620150829).
 Bernhardt, T., Kanay De, S., Thida, M.W. and Min, A.M. (2016) ‘Myanmar’s new minimum wage: What’s next? Policy considerations for the way forward’. CESD Labor Market Reform Working Paper No. 1/2016.
 Bernhardt T., Kanay De, S. and Thida, M.W. (2017) Myanmar labour issues from the perspective of enterprises: Findings from a survey among food processing and garment manufacturing enterprises, ILO, CESD, GIZ.
 Oxfam (2015) ‘Made in Myanmar: Entrenched poverty or decent jobs for garment workers?’. Oxfam Briefing paper no. 209.
 Tudor, O. (2015) ‘Burma: Unions, global brands and NGOs back minimum wage for all’. Stronger Unions, 16 July (http://strongerunions.org/2015/07/16/burma-unions-global-brands-ngos-back-minimum-wage-for-all/).
 Tanaka, M. (2017) ‘Exporting Sweatshops? Evidence from Myanmar’ (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2938903).
 Raitzer et al. (2015) Myanmar’s Agriculture Sector: Unlocking the Potential for Inclusive Growth. ADB Economics Working Paper Series, No. 470.
Photo credit: NYU Stern BHR via Flickr