Karachi is a Small City

By Anjum Altaf

City size is back in fashion as a variable of interest and this time bigness is being viewed as an advantage. This is quite a change from the perspective that prevailed for years when countries, specially developing ones, were decidedly anti-urban and wished to retard migration to prevent cities from increasing in size. Size was seen as a handicap and served as an excuse to explain away the problems of big cities. How should we see Karachi in this new perspective?

Of course, well-managed big cities have been around for a long time – Tokyo, New York and London are obvious examples. But somehow it was felt that such success could not be replicated in developing countries. The blame was always placed at the door of mismanagement though it was never adequately explained why such mismanagement was so endemic to developing countries and why the ability to manage declined with size. Now, following the East Asian miracle, the emergence of dynamic big cities like Seoul and Shanghai has re-opened the debate about the possibilities and dividends of urban size.

Recent research has in fact suggested that most Chinese cities are still too small from an economic perspective and that it is much more costly to be undersized than to be oversized. The reason is that even after accounting for the negative effects of size like congestion, productivity, or output per worker, increases significantly with size.

We might choose to ignore these potential economic gains if small cities were exceptionally well managed in Pakistan. But given that this is not the case by any stretch of the imagination, there seems no reason to fail to tap the gains available from increases in city size.

To do that, however, calls for a re-examination of the reasons for the poor performance of large cities in many developing countries in the world. Here, I feel, the hand-wringing about mismanagement is an easy cop-out that hinders the search for what might be considerably more significant reasons.

In this perspective I wish to advance the claim that size and effective size are very different things and while Karachi is large in size, is effective size is quite small. And this divergence between size and effective size has serious implications.

The key to understanding this difference is the realization that productivity, or output per worker, increases with city size because of the resulting increase in the size of the labor market. Not only is there a greater diversity in the labor skills available to firms but the greater number of available jobs enables labor to match itself with the job that provides the highest returns.

But, and it is a big but, this requires that it be physically possible for the optimal match to occur. If one resides at one end of Karachi and the best paying job exists at the other, it should be possible to travel to and from the location of the job in a reasonable amount of time and at a reasonable cost. In other words, the benefits of size in effect derive from the existence of an integrated labor market. It is in this sense that Karachi is not a city of 18 million people; it is perhaps akin to six cities of three million people each. It suffers all the social and political disadvantages of a large population without deriving any of the economic benefits.

How can one understand the concept of effective size? It turns out that the indicator is a very simple one which can be illustrated with an example. Let us consider one hour as the maximum feasible time for a home-to-job commute and two hours as the maximum feasible time for firms making just-in-time deliveries to other firms (for example, bakeries delivering fresh products to hotels) . Between 1990 and 2007 the population residing within the one and two hour circles from Shanghai port increased 200 percent (from 4 to 12 million) and 130 percent (form 10 to 23 million), respectively. These improvements came from investments in transport infrastructure that integrated the labor market greatly reducing the divergence between size and effective size. And such investments in China were not confined to Shanghai; smaller cities like Kunshan and Suzhou showed even higher gains. Not surprisingly, the economic output per square kilometer increased between two and three fold within a period of five to ten years.

The bottom line of this argument is that contrary to our belief, Karachi is not a big city; it is an ineffective agglomeration of about half a dozen small cities whose labor markets need to be integrated to avail the benefits that derive from bigness. This requires investments in transport infrastructure, rapid transit, and the sensible utilization of the circular railway that has been criminally neglected over the years.

One of the unremarked advantages of Mumbai is its suburban railway system that transports about 7 million commuters per day. It is not surprising that in 2008 the gross domestic product of Mumbai was $209 billion compared to $78 billion for Karachi. It is not that Mumbai is significantly better managed or has fewer social and political problems. A large part of the difference pertains to the integration of its labor market. In terms of size Mumbai and Karachi are comparable. In terms of effective size, Karachi is much smaller, perhaps one-third the size of Mumbai.

The urban agenda for Karachi and the other major cities in Pakistan couldn’t get much clearer than this. In fact, looking ahead there should be no reason why the labor markets of Hyderabad-Karachi, Gujranwala-Lahore or Islamabad-Rawalpindi should remain unconnected to each other.

This Op-Ed appeared in Dawn, Karachi, on November 15, 2010 and is reproduced here with the author’s permission to facilitate a discussion.


  • a ercelan
    Posted at 16:34h, 16 November Reply

    Ignoring the author’s wish to disarm critics by enclosing himself and the reader in an “economic perspective,” this note exhibits an unfortunately common tendency to downplay ecological fundamentals that go beyond congestion and the like. Yes, Karachi is quite likely to reduce income poverty and perhaps even provide prosperity to immigrants. Yes, hope is a striking counterpoint to fear as evidenced most recently and powerfully by the number of IDPs – both men and women – who are reluctant to return to landlessness and associated oppression. But the city is literally choking on its pollution, an inevitable consequence of size. Frictions and fragmentation are so enormous that the city is hostage to endemic violence. Mismanagement is often an illusion – Karachi’s political elite are very effectively managing their own fortunes.

    It is fashionable to push for rapid transit. But why should people have to find decent work at costly distances from their homes? How can you build caring communities under such disconnect?

    As to Pindi-Isld integration of labour markets, my impression is that this is already happening – in fact as distantly as murree and gujar khan!

    • Vinod
      Posted at 03:20h, 17 November Reply

      If ercelan is correct about the ecological problems of Karachi, then it will affect the productivity of the city even if the article’s point about connecting labour markets is achieved. But the former problem should not stop the city from trying to achieve the benefits of the latter solution. While they are linked they are not inextricably dependent.

    • Anjum Altaf
      Posted at 05:57h, 17 November Reply

      ercelan: Vinod has already pointed to the problem in your comment. There are other points that need clarification and I will mention some of them:

      1. What is the “economic perspective”?
      2. What are the ecological fundamentals?
      3. If pollution is an inevitable consequence of size, Seoul and Shanghai should be drowning in pollution. They are not. Conversely, Sialkot should have much less pollution than Karachi. Does it? What is the measure of pollution you are using? In general, small industrial cities are more polluted than large post-industrial cities.
      4. How is the endemic violence ameliorated by not having efficient public transit?
      5. The point of public transit is that the home-to-job commute is not costly. Do you have any example of a caring community where everyone has a decent job where they live? When people marry, would they have to stipulate that the spouse have a job in the same community? Without public transit Karachi is six small cities juxtaposed to each other and not one large efficient city. Is this to be considered good because it yields caring communities?
      6. The Pindi-Islamabad labor market is far from integrated. A person living in Pindi without personal transport cannot turn up for work in Islamabad every day at 8 am unless he starts at 5 am. That is the reason why the one-hour commute indicator is used to define an integrated labor market.

  • Anjum Altaf
    Posted at 23:11h, 20 July Reply

    Once again, some out-of-the-box thinking on urbanization and city structure from China:


    Note the reference to the one-hour commute circle design parameter which we used in our research on Lahore:

    “Chinese planners used to follow a rule of thumb they learned from the West: All parts of an urban area should be within 60 miles of each other, or the average amount of highway that can be covered in an hour of driving. Beyond that, people cannot effectively commute.”

    Also note the vision of linking small cities with the metropolis in a regional economy. For a comparison, see the policy brief of our work on Lahore: http://cdpr.org.pk/…/upl…/2014/10/Metropolitan-Lahore-PB.pdf

Post A Comment