NBER Working Paper Discussion | Growing, Shrinking, and Long Run Economic Performance: Historical Perspectives on Economic Development


This paper by Stephen Broadberry and John Joseph Wallis attempts to shift focus from the typically discussed topic of economic growth, to the often-neglected topic of economic shrinks. The idea is that controlling economic shrinks has more of an effect on long-run economic growth, than promoting economic growths. To test this theory, Broadberry and Wallis examine historical data from multiple time periods, to find trends in the growing and shrinking rates of poor, middle-income, and high income countries.

The terms “shrink” and “growth” refer to instances where a country’s economy gets smaller or larger. If, over the course of the year, the economy (measured by GDP per capita) gets smaller, we call this a shrink. Conversely, if we observe the economy to be getting larger, we call it a period of growth. In theory, if a country were to maximize its periods of growth, it would see long-term sustained growth. And in fact, much of the current research into growth theory has been about how to maximize the frequency and magnitude of growth periods. The information that Broadberry and Wallis present shows that the scope of this type of research is not wide enough, and that economic shrinks have a substantial effect on a growing economy, despite the lack of attention they have gotten.

Methods and Findings

To arrive at this conclusion, Broadberry and Wallis create values that they call “contribution of shrinking” and “contribution of growing” which essentially equal the frequency of shrinking times the rate of shrinking, and the frequency of growing times the rate of growing, respectively. Here, the rate of change refers to the average of all the shrinking or growing years over the stated period. Looking at short-run growth data, from 1950-2008, and long-run growth data, from 1348-2008, broken down into shorter periods of about 50 years, Broadberry and Wallis arrive at a few specific conclusions. First, growing and shrinking rates have been high and variable throughout most of history, and remain so in less developed countries today. This finding indicates that more developed countries have found a way to temper not only shrinking, but growing as well, and that consistency in rates of change is key for economic success. Second, improving long run economic performance has occurred because the frequency and rate of shrinking have both declined, rather than because the growing rate has increased. This may be the most important conclusion drawn by Broadberry and Wallis, because it provides a new metric for developing countries to target with policies. If, previously, countries were focused on achieving high growth, and neglecting shrinks, this finding represents a paradigm shift, and one that should, in theory, see results. Finally, the authors conclude that the rate of growing has typically declined rather than increased as long run economic performance has improved. This is interesting, because it is nearly antithetical with what one would intuitively suspect regarding growing rates.

After discussing the empirical evidence, Broadberry and Wallis then go on to mention how countries might stop their economies from shrinking, discussing both proximate and ultimate causes of economic shrinks. Four possible proximate causes are given to explain economic shrinks: structural change, technological change, demographic change, and changing incidence of warfare.

Structural change describes the shifting of an economy over time from being centered in one industry to deriving most of its growth from another. The paper uses Great Britain as an example, citing the shift from an agrarian economy in the pre-1800s, to one centered around services and industry (or manufacturing) now. The authors contend that developments within the services and industry sectors increased the net growth rate of Great Britain substantially.

Technological change by itself, in theory would lead to an upswing in the GDP of a nation. For this reason, Broadberry and Wallis chose to observe a metric called Total Factor Productivity, which roughly describes how efficiently inputs are used in production. An increase in TFP is associated with increasing efficiency in production, and is thereby used to describe a positive change in the level of technology. By looking at long-run growth rates in both TFP and output, Broadberry and Wallis found that economic growth outpaced technological change, between the formative years of 1760-1873, so that tech change only accounted for just one sixth of the increase in growth. They also found that a decrease in TFP coincided with a shrink in Holland in the 1700’s.

Demographic change is a field of study popularized by Thomas Malthus in the late 16th/early 17th centuries. Malthusian theory of population growth says that since population growth is geometric, and resource growth is linear, society will get to a point where a population will outgrow its resources, and experience short run shrinking. In this approach, any measure that increases birthrate, or decreases available land (resource constraint) will result in a short-run shrink (and vice-versa for growth). Broadberry and Wallis note that, for developing economies, this theory holds true.

Finally, changing incidence of warfare is just that – the increase or decrease a country sees in the incidence of warring over a period. Economic theory can be found supporting either side of this concept with regards to growth. Warfare can bolster a country’s manufacturing sector, and increase investment, while also increasing the mortality rate – all things that should lead to economic growth. Conversely, warfare necessarily leads to a decrease in the labor force, a possible decimation of infrastructure and physical capital, and a larger segment of the population dependent on social safety nets. What Broadberry and Wallis find, is that incidence of warfare is associated with variable growth and shrink rates.


These four proximate factors are found to be important by themselves in reducing economic shrinks, which leads Broadberry and Wallis to consider the ultimate factor of institutional change. The authors divide a political system into two different possible areas, one based on what they call “identity rules” and one based on “impersonal rules”. In short, an economy is more likely to experience less frequent and less severe shrinking periods once they transition from identity rules system of government, where different rules apply to different levels of society, to impersonal rules, where the same rules apply for everyone in society. Broadberry and Wallis then go on to mention how a government acting within an impersonal rules system can operate though changes in the proximate factors, to bring about periods of sustained growth through limited periods of shrinking. Key to this theory is a move away from agriculture, towards technological progress, and towards a more peaceful environment with stable demographics.

This paper refers directly to the development of nations, and as such it can be used as a benchmark for emerging economies. Since much of the research and theory being done now is on growth factors, this paper, and the conclusions that it draws – specifically the importance of limiting frequency and magnitude of shrinks – provides a new perspective on an age-old question. In addition, many poor and developing countries in the world now have political systems that can be described as “identity rules”. It is likely, according to the research done in this paper, that this one aspect may be holding these economies back and preventing them from making the transition into sustained growth.

By: Garrett Blom


For reference, the paper can be found here: http://nber.org/papers/w23343


4 thoughts on “NBER Working Paper Discussion | Growing, Shrinking, and Long Run Economic Performance: Historical Perspectives on Economic Development”

  1. This post does an excellent job of simplifying the results of this paper into an easily-digestible format. It is well-organized and interesting. The results of this paper contradict expectations; most would anticipate that promoting growth would be the primary actor in improving economic growth. The fact that this paper has found that preventing shrinks to be more important is very intriguing. I intend to look into this more deeply in the near future.


  2. Very well written post. You did a great job of introducing what you were talking about for the author of somebody who has not taken this class. You gave an accurate overall scope of economic development from the paperand how it is analyzed from each perspective. This provides a unique intuition for economic growth in developing countries and how history has impacted their reasonings. The goal is to reduce economic shrinkage and volatility and that will increase productivity and GDP per capita.


  3. This is a new and interesting take on ideas of how to develop countries. I can definitely see how the “Identity Rules” governance applies in places like India and the Middle East. Given the instability and essentially continuous warfare in places like Africa and the Middle East, I really wonder if those regions can ever converge with Westernized countries, just due to the strains of it all and the frail state it puts those economies in.
    Also, given that demographics is a key factor you mentioned, it makes me wonder about the countries in Europe who are bringing in so many migrants, a majority from developing countries. It’ll be interesting to see if they
    ‘revert’ and experience more frequent and intense shrinkage.


  4. This was a really interesting article. It is interesting to see additional pieces to the theory of economic growth than the Solow Model.
    I remember talking about institutional strength as a potentially contributing factor to economic growth. Specifically, if a nation was based on value creation or resource extraction. This paper relates to this idea in a way. Strong institutions can help to mitigate the cycle of the economy; prolonging the growth periods and reducing the shrinking periods.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s