Sunday, November 23, 2014

The Growing Divide Within Developing Economies

PRINCETON – When researchers at the McKinsey Global Institute (MGI) recently dug into the details of Mexico’s lagging economic performance, they made a remarkable discovery: an unexpectedly large gap in productivity growth between large and small firms. From 1999 to 2009, labor productivity had risen by a respectable 5.8% per year in large firms with 500 or more employees. In small firms with ten or fewer employees, by contrast, labor productivity growth had declined at an annual rate of 6.5%.

Moreover, the share of employment in these small firms, already at a high level, had increased from 39% to 42% over this period. In view of the huge gulf separating what the authors called the “two Mexicos,” it is no wonder that the economy performed so poorly overall. As rapidly as the large, modern firms improved, through investments in technology and skills, the economy was dragged down by its unproductive small firms.

This may seem like an anomaly, but it is in fact an increasingly common occurrence. Look around the developing world, and you will see a bewildering fissure opening up between economies’ leading and lagging sectors.

What is new is not that some firms and industries are substantially closer to the global productivity frontier than others. Productive heterogeneity – or what development economists used to call economic dualism – has always been a central feature of low-income societies. What is new – and distressing – is that developing economies’ low-productivity segments are not shrinking; on the contrary, in many cases, they are expanding.

Typically, economic development occurs as workers and farmers move from traditional, low-productivity sectors (such as agriculture and petty services) to modern factory work and services. As this takes place, two things happen. First, the economy’s overall productivity increases, because more of its labor force becomes employed in modern sectors. Second, the productivity gap between the traditional and modern parts of the economy shrinks, and dualism gradually diminishes. Agricultural productivity increases during this process, owing to better farming techniques and a decline in the number of farmers working the land.

This was the classic pattern of postwar development in the European periphery – countries like Spain and Portugal. It was also the mechanism that generated the Asian growth “miracles” in South Korea, Taiwan, and eventually China (the most phenomenal example of all).

One thing that all of these high-growth episodes had in common was rapid industrialization. Expansion of modern manufacturing drove growth even in countries that relied mostly on the domestic market, as Brazil, Mexico, and Turkey did until the 1980’s. It was structural change that mattered, not international trade per se.

Today, the picture is very different. Even in countries that are doing well, industrialization is running out of steam much faster than it did in previous episodes of catch-up growth – a phenomenon that I have called premature deindustrialization. Though young people are still flocking to the cities from the countryside, they end up not in factories but mostly in informal, low-productivity services.

Indeed, structural change has become increasingly perverse: from manufacturing to services (prematurely), tradable to non-tradable activities, organized sectors to informality, modern to traditional firms, and medium-size and large firms to small firms. Quantitative studies show that such patterns of structural change are exerting a substantial drag on economic growth in Latin America, Africa, and in many Asian countries.

There are two ways to close the gap between leading and lagging parts of the economy. One is to enable small and microenterprises to grow, enter the formal economy, and become more productive, all of which requires removing many barriers. The informal and traditional parts of the economy are typically not well served by government services and infrastructure, for example, and they are cut off from global markets, have little access to finance, and are filled by workers and managers with low skills and education.

Even though many governments exert considerable effort to empower their small enterprises, successful cases are rare. Support for small enterprises often serves social-policy goals – sustaining the incomes of the economy’s poorest and most excluded workers – instead of stimulating output and productivity growth.

The second strategy is to enlarge opportunities for modern, well-established firms so that they can expand and employ the workers that would otherwise end up in less productive parts of the economy. This may well be the more effective path.

Studies show that few successful businesses begin as small, informal firms; they are started, instead, at a fairly large scale, by entrepreneurs who pick up their skills and market knowledge in the more advanced parts of the economy. Enterprise surveys in Africa by John Sutton of the London School of Economics indicate that it is often entrepreneurs with experience in importing activities who found modern domestic firms. Domestic subsidiaries of multinational firms or state-owned enterprises – which are repositories of skilled workers and managers – are also a source of such firms.

The challenge is to create an economic environment in which there are incentives for local talent and capital to invest in firms in the modern, tradable parts of the economy. Sometimes, it is enough to remove certain of the more stifling government regulations and restrictions. At other times, governments need more proactive strategies – such as tax incentives, special investment zones, or hyper-competitive currencies – to raise the profitability of such investments.

The details of appropriate policies will depend, as usual, on local constraints and opportunities. But every government needs to ask itself whether it is doing enough to support the expansion of capacity in the modern sectors that have the greatest potential to absorb workers from the rest of the economy.

Read more from "Submerging Markets?"

  • Contact us to secure rights


  • Hide Comments Hide Comments Read Comments (11)

    Please login or register to post a comment

    1. CommentedEnrique Woll Battistini

      Entropy has a way of manifesting itself in every corner of reality... In all activities. The economy is not excepted. In this case, the mechanics of this phenomenon, of increasing productivity in larger firms and decreasing productivity in smaller firms is likely the result of simple economies of scale, where larger firms can afford the newest most expensive technologies required for increased competitiveness in the goods and services demanded by the market, and the smaller ones cannot.

    2. CommentedArjan de

      Dani, while this is an important point, I think the analysis needs to be sharpened. This divide has always existed, at least since the colonial economist Boeke described it, and very much since the re-discovery of the informal sector in the late 196os in Kenya. I am also not sure what the 'typical' economic development path means (least of all China, where agric prod increased first, anyway China is atypical in this respect). But that’s all splitting academic hairs; what’s really necessary is to think more about recommendation that the formal sector need support, and the dualism in that thinking, which arguable is one of the main reasons the informal sector does remain where it is (it’s not a ‘traditional’ sector, and it is ‘informal’ because policy makers and analysts say so).

    3. Commentedde Lafayette

      {What is new – and distressing – is that developing economies’ low-productivity segments are not shrinking; on the contrary, in many cases, they are expanding. }

      Once again, we must look at how we define productivity. There seems to be multiple definitions. Productivity is a ratio of output over some input factor - often either hours worked or labor-costs.

      Frankly, the one I like most is Multi-factor Productivity that indicates, finally, neither hours worked nor labor-costs alone are key in the defining the ratio.

      I can think of another, far less amenable to statistical reduction. It is the fact that Labor mobility in a country like the US is high throughout. But not as high in Europe, where only a tiny percentage of the population speaks a foreign language, thus reducing the propensity for mobility.

      The OECD is taking a major step in the right direction in defining multi-factor productivity. The complexity of productivity is amazing - I highly suggest one have look at a compendium of the OECD's studies here:

    4. CommentedLuc Lapointe

      Dear Dani...interesting article but I am not sure where you saw such low employment and high demand for this sector that is associated with low-skills. Correct me if I am wrong but I haven't seen meaningful efforts to create substantial employment for this sector. The only one that would benefit would be the government for collecting more taxes from the poorest.

      You may have noticed that the new "stars" of development are Unilver, JP Morgan Chase, Coca Cola, to name a few. Would these be job creators? Would dumping cheap "soft drink", "Ragu" and "Lifebuoy" in developing markets a good thing?

      Most developing countries have experienced rapid growth during the recent golden age but failed miserably to do this is a sustainable way. As the world economies slow down, developing countries will face sluggish growth and having to deal with crumbling infrastructures.

      Moving into the formal economies will not help the poorest. The services available to them (education, health, finance) lack quality. The system is so disconnected with the reality - the policies to reduce extreme poverty could be compared to waste management in developed countries - if you make them invisible people think that the problem is resolved. It is a false assumption to believe that there are more doctors and teachers or that the government will support them for the short run with finance and better services.

      ....but in a perfect world...I agree - it should work!

    5. CommentedBrett Blake

      As for the "second strategy," my working assumption right now is that global-facing companies will be very selective and slow to increase employment for the time being. The thinking seems to be tight control on payrolls and exercising great care in the capital investment these high-productivity jobs require. This could change, but I see no signs on the horizon of this happening in any but a fitful manner.

    6. CommentedHamid Rizvi

      I did not see any mention of a comparison of the productivity figures which according to you perhaps or MGI was a “remarkable discovery”. It would be telling to see what the productivity figures were for instance the last 10 and even 20 years prior. These figures may suggest the new numbers are not all that remarkable after all.
      As someone has already pointed out and I would tend to agree there are a number of factors at play. Like:
      • The economies of improving scales over time (available to large entities)
      • Technology the catalyst for the first
      • Better more sophisticated Management of resources
      • Accounting practices are a major factor
      • Public private partnership. A relationship that needs to be handled with utter sensitivity not to disturb natural market balance.
      • The Walmart phenomenon
      To name a few. Small businesses have traditionally fared poor in productivity as compared to large enterprises for reasons that I cannot intelligently enumerate. The important thing to remember is and one that you correctly pointed out is the changing mechanism of going from small to large is different now then it traditionally was. How this will play out in the next say 2 decades will be interesting to see. Perhaps, MGI, will help us with better figures there. It’s a tricky proposition in my opinion that can upset natural market mechanisms and create something similar to the tech bubble. Giants with hollow insides. On a bigger scale India comes to mind. I call India the Microsoft Economy. In this case restarting will not help any!

    7. CommentedProcyon Mukherjee

      Dani, your paper clearly brings out the allocative inefficiency that creeps in while structural shift happens as economies diversify away from either agriculture to industry or from small scale to large scale industrialization.

      But I was pleasantly surprised to see one country handle this far better, Vietnam, where diversification of agriculture to processed food manufacturing and animal husbandry was handled so well. Here perhaps the invisible hand is far less effective and the allocative power of the state did far better.

    8. CommentedDaryl stevens

      Premature De-industrialiation

      This is largely do to the Trade Diversionary Models that exist in some regions of the world, and toward which, because of its existence, we will see more movement.

      Unfortunately, the countries which have benefited, developed, are moving through development are subservient to their domestic elites, and impact global trade, have global impacts which will see movement that counters their policies, especially after the world stabilizes; which it has been doing. This is the big story, while pundit after pundit discusses the Rise of the Rest, or the Transfer of power from X to Z, they are missing the real story, that is how the "insuring against financial instability" a la Feldstein, otherwise known as, "the US is reliant on Foreign Funding", and the policies that attract investment are trade diversionary, not trade creative. Where some have gone to such extent, bloating assets, printing money, and running up savings, they impair the ability of trade to move, for production to move, blowing up capacity globally, and creating such vast imbalances, that it diverts the capacity and trade flows, investment flows that would proliferate more globally. this is pushing, many small countries, even ones at very low levels of development into a post-industrial society scenario, before they even develop, as elites in some regions push asset bloat, and investment to un-useful heights, rather than switch to consumer markets, and free up their societies for education, training and services.

      The emerging scenario is far less advanced country, developing country, than it is developing country developing country, as it has always been, but for the preference of some elites to see global equality in their remote control lifetime.

    9. CommentedJose araujo

      And alas, Dany discovered what economies of scale are...

      He needed a Mckinsey study to figure the impact of size on productivity...We need to get him a copy of Sammuelson or some economics 101 text book

        CommentedJose araujo

        Brian, if companies are getting larger, others are getting smaller thats how economies of scale work, leading in extreme situations to monopolies.

        This is how industrialization and development is made. The main determinants of productivity are

        1 - Scale
        2 - Technology

        All other stuff is political demagogy and prejudices

        CommentedBrian Warby

        Economies of scale would only explain why large firms are more productive than small firms. It definitely does not explain why small firms are becoming less productive than they have been in the past.

        The real question is, what's driving people to the smaller firms? If the large firms are more productive, they should be able to pay more. So, is it high entry barriers? Perhaps more people are moving from the informal to the formal economy? An abundance of labor and shortage of expertise in the productive industries?

        This seems rather problematic considering the expansion of microfinance and efforts to build small and medium enterprises.