Labour Productivity In Pakistan: Why Are We Falling Behind?

This article is part of a series titled “Is there a way forward for Pakistan?” Read more about the series here.

The scale of the economic challenge we as a nation face is best summarised by Figure 1. While labour productivity in Pakistan only increased by 45%, labour productivity in Bangladesh, India and China (not reported) saw an increase of 190%, 263%, and 790%, respectively, over the last three decades. What is more revealing is that, compared to these countries, the increase in labour productivity has been the lowest across almost all the sectors in the case of Pakistan.

An obvious question that comes to mind is why has Pakistan lagged behind relative to its neighbouring countries? And what can be done to reverse this trend?

To better understand why Pakistan has lagged behind its regional peers in terms of labour productivity, it is important to start by recognising that labour productivity depends on both the level of productivity and the physical capital available for production. This distinction between the level of overall productivity and the physical capital as two driving factors underlying labour productivity is of critical importance. The popular discourse on labour productivity often ignores the role of physical capital and confuses it with productivity in general. This is a mistake.

In the rest of this article, I discuss both the role of productivity and physical capital to explain why labour productivity in Pakistan has increased by so little since 1990s.

In an exercise done by this author, a one percent increase in productivity can potentially increase Pakistan’s GDP by 2.53%. This is significantly higher than what Jones (2011) estimates for most other economies. However, the overall productivity growth in Pakistan has remained abysmal to say the least.

Figure 2 plots the annual productivity growth for Pakistan for the period between 1968 and 2018. Since 1968, the average productivity growth is estimated to equal 1.24%. It was only for the short period in the 1980s that the average productivity growth reached close to 3%. The 1990s and the 2000s once again saw the annual productivity growth fall below 1%. It then increased to 2.5% during the recent decade. The key message that comes out of this exercise is that, despite significant benefits in terms of economic prosperity, the productivity growth rate has continued to fall short of what is required to sustain long periods of high economic growth.

But once again we are forced to ask a question, why has productivity growth remained low in the case of Pakistan? In one of his lectures, Charles Jones notes, “Poor countries are poor partly because of few inputs but also because of inefficiency in using those inputs.” In a similar spirit, Franklin Fisher had earlier remarked, “In dealing with actual economies, the barriers (for resources to move) may be more important than the frontier.” It is easier to appreciate this if one considers that, over the past five decades, almost 40% of the growth in GDP per person for the United States came from the better use of its human resource i.e. right person for the right job. The potential benefits for developing economies from improving how they use their existing resources are also enormous.

To better understand why Pakistan has lagged behind its regional peers in terms of labour productivity, it is important to start by recognising that labour productivity depends on both the level of productivity and the physical capital available for production.

Conceptually, the poor use of resources affects a country’s development prospects by undermining overall productivity. For example, Hsieh and Klenow (2009) show that inefficient utilisation of resources may reduce productivity by about two to three times in the case of China and India. In its vision document, the Economic Advisory Group cites several examples from the literature showing how several of the developing countries improved their overall productivity by letting their existing resources move to more productive activities. Meza et al. (2019) show that 41% of the increase in Mexico’s productivity between 2003 and 2012 was because of improvement in allocative efficiency, i.e. due to economic resources moving from less productive to more productive activities. Improvement in allocative efficiency was an important part of productivity growth during Chile’s decade-long period of growth following the debt crisis of the early 1980s (Chen and Irrazabal, 2015).

At the onset of the 2013 crisis, organised sectors and businesses linked to powerful families successfully lobbied to increase trade protection in the form of non-tariff measures to protect themselves from international competition.

In the context of Pakistan, the two broad themes which have the potential to cover much of the ground on inefficient use of economic resources are trade protection and the low levels of labour mobility. Pakistan undertook significant trade liberalisation during the 1990s and the 2000s. The tariff rate came down from more than 50% in the 1990s to less than 20% in the late 2000s (Malik and Duncan, 2022). The authors show that much of this progress was partly reversed during the last decade. However, while the focus on the average tariff rate gives useful information on the degree of protection from international competition, it hides important sources of distortions. A closer look at the data shows that the effective rate of protection enjoyed by the sectors dominating Pakistan’s economy is substantially higher (Varela et al., 2020). It is this excessive protection of traditional sectors from international competition which prevents Pakistan’s economy from adapting to the requirements of the 21st-century economy.

The second theme centres around low levels of labour mobility. Consider the fact that only 15% of the total migration that happens within Pakistan happens for economic reasons (LFS, 2021). Understanding why this is the case requires asking if this is due to a lack of affordable housing; poor transportation network; issues around security of life and property; dependence on social networks due to unreliable provision of public goods such as health care; Or lack of opportunities. Pakistan, as a country with a relatively abundant labour force, has enormous potential in competing globally in products which use labour more intensively in production. But if the country’s labour force cannot relocate in response to the needs of the economy, the potential benefits from integrating with the rest of the world will not materialise.

The two themes discussed above leave much else to be desired. Some of these include liberalising pricing regimes and replacing minimum support prices with instruments such as crop insurance for small farmers; revamping the education system with the aim to introduce and mainstream pathways for vocational training at the level of higher and post-secondary education; designing appropriate tax policy to discourage speculative investments in both urban and farmland; undertaking judicial and civil service reforms; and, importantly, the democratisation of political parties (EAG, 2020).

However, addressing the structural issues mentioned above is often not straightforward for both political and technical reasons. On political reasons, Jones points to the economic interests of the ruling elite as an important factor behind why a country’s resources are not used efficiently. Jones (2013) says, “The state-of-the-art in that literature suggests that misallocation is the equilibrium outcome of a political process interacting with institutions and the distribution of resources (including physical capital, human capital, ideas, and natural resources). It is, evidently, not in the economic interest of the ruling elite to improve the allocation of resources, despite the potentially enormous increase in the size of the economic pie that is possible in the long run.”

In a recent paper published by PIDE, Adeel Malik and William Duncan document this phenomenon in the context of Pakistan. They show how at the onset of the 2013 crisis, organised sectors and businesses linked to powerful families successfully lobbied to increase trade protection in the form of non-tariff measures to protect themselves from international competition. Likewise, the 2018 crisis saw a sharp increase in import duties in sectors linked to powerful families. The 2022 crisis has proven to be no different.

On the technical side, Fisher reminds us, “It is very risky to look at a planned economy and predict where it will end up if it shifts to a system of free markets.” Likewise, Yao (2014) says, “High growth is not granted when a command economy is transformed into a market economy.”

However, these challenges are not specific to Pakistan alone. As low as it might first appear, the average productivity growth in Pakistan is not too different from what is observed for the rest of the world. For example, since 1980, the average annual productivity growth in the case of India has been 1.5%. The same for Pakistan has been 1.4% over the same time period. In fact, Pakistan’s average productivity growth has been three times higher than the rest of the world. Figure 3 plots the distribution of annual productivity growth for the country-year pair. There are two important points to note. First, there is considerable variation in the annual productivity growth across country-year. Second, the average annual productivity growth across country-year has only been 0.4%.

A cursory look at the data also suggests that there is no obvious relationship between the level of economic development and productivity growth. The correlation between annual productivity growth and the level of GDP is close to zero. In other words, it is not the case that countries at any level of economic development experience faster productivity growth on average.

This observation is not new. There is already considerable literature showing that the contribution of productivity growth to the East Asian growth miracle was rather unimpressive. For example, Collins and Bosworth (1996) estimate that, while GDP per capita grew at an average rate of 4.2% over the period from 1960 to 1994, the contribution of productivity growth to the annual GDP per capita growth rate was only 1.1 percentage points on average.

Admittedly, the above discussion should not be taken as conclusive. There are countries which succeeded in achieving high growth rates driven by continuous improvement in productivity. For example, output per hour increased at an average rate of 3.3% in the US between 1948 and 1973. Jones (2016) shows that almost all of it can potentially be explained by advances in productivity.

Moreover, methodological choices can also change results substantially (Barro and Sala-I-Martin, 2004). For example, taking a different approach, Hsieh (2002) shows that the contribution of productivity growth to annual GDP growth for Taiwan increases from 2.1 percentage point to 3.7 percentage point. The same for Singapore increases from close to zero to 2.2 percentage point.

The more relevant question in the context of this article is how else can the countries sustain long periods of high economic growth. After all, countries across the world have made substantial gains when it comes to improving the living standards of their citizen without a substantial increase in their overall productivity. This provides the motivation for understanding the role of physical capital as the second key driver of improvements in labour productivity.

It is argued that for countries which are broadly similar, the country with less physical capital will offer higher returns on investment. As a result, if these countries were to open up their economies to foreign investment, they will experience an inflow of capital which will contribute to their economic development. While this is not always the case, this is certainly true for several emerging economies, including India and China.

Figure 5 shows that, as India and China opened up their economies to foreign investment in the early 1990s, they experienced a significant increase in inflows from international private investors (blue bars).

In contrast, while Pakistan has also seen investment from international private investors, these inflows have not sustained for longer. In effect, most of the non-debt-creating inflows are concentrated in a few years during the last three decades. Figure 6 shows that the only time net private equity inflows remained above 2% of GDP for more than a year was in the mid-2000s.

By not accumulating reserves, policymakers have effectively prioritised today’s consumption over future consumption. Indeed, in doing so, they have effectively left the citizens worse off.

In addition to foreign investment, domestic investment in these countries has increased as well. For example, the gross savings rate (as % of GDP) increased from close to 15% during the 1980s to more than 30% in the 2000s.

China and Bangladesh have also seen a significant increase in their savings rate during this period, except that the increase for China is smaller due to an already high savings rate, to begin with. In contrast, the gross savings rate for Pakistan has declined over the same period.

Why is it that both foreign and domestic investment has not increased in Pakistan despite Pakistan having comparable socio-economic conditions to the rest of the countries in the region? Like always, there is no one good answer to a question as important as this. Nonetheless, I put forward the reason which I think is most important: macroeconomic risk in the form of frequent currency crises.

The best way to visualise how macroeconomic risk may affect firms’ investment decisions is through Figure 7. Figure 7 plots the annual return on a KSE100 five-year index fund. I select five years since it matches the average duration of the business cycle and, as a result, is most suited to capture macroeconomic risk. Two things stand out. First, the average return on this index fund equals 16%. In real terms, the average return equals 8%. Second, the standard deviation (i.e., risk) is also quite high at 17%.

The two observations are of critical importance. The higher average returns are consistent with the notion that less developed countries tend to offer higher returns on investment. As a result, liberalising the economy should result in an increase in investment. However, there is also considerable macroeconomic risk associated with undertaking investment in Pakistan. The average return of 16% and standard deviation of 17% means that the Sharpe ratio is less than one, thus making both foreign and domestic investment less attractive.

This simple observation has important policy implications. A better set of macroeconomic policies which lower the risk of repeated currency crises can go a long way when it comes to incentivising investment and bringing about substantial improvements in labour productivity. The set of macroeconomic policies which help achieve this is well known. However, in the context of Pakistan, the role of accumulating foreign reserves deserves additional attention.

Figure 8 highlights an important difference between Pakistan and the regional economies discussed above. While all the economies received a varying degree of private inflows over the four decades, all the countries except for Pakistan used these inflows to accumulate foreign reserves. In fact, China accumulated more in foreign reserves than the net private inflows it was receiving in any given year for more than a decade. It is easy to appreciate that the macroeconomic risk in developing countries with large reserve cover is definitely less than in the case of countries which don’t.

In the discussion above, I have stepped back and taken a broad view of both the structural and macroeconomic challenges Pakistan face. Without going into technical details, I have also hinted at several of the reform measures the policymakers can take to address these challenges and lead Pakistan’s economy out of the quagmire it is stuck in.

However, the discussion skips several important questions which are arguably more fundamental to understanding the challenges highlighted above. For example, by not accumulating reserves, policymakers have effectively prioritised today’s consumption over future consumption. Indeed, in doing so, they have effectively left the citizens worse off. But why is it the case that policymakers in Pakistan prioritise today’s consumption more than policymakers in our neighbouring countries? Second, while I am equally guilty of doing this above to get my arguments across, there is no good reason to believe that what is ‘macroeconomic’ is independent of what is ‘structural.’ It is very much likely that the structural problems which prevent sustained growth also force policymakers to bet on expansionary (procyclical) policies to deliver short periods of growth for electoral victory. But, if so, what prevents the politicians from addressing these structural bottlenecks that ultimately give way to irresponsible macroeconomic policies? I leave it to the reader to explore the answers to these questions.

Ahmed Pirzada

The writer is Chairperson of Economic Advisory Group, Pakistan; Senior Lecturer in Economics,
University of Bristol; and, Fellow of Higher Education Academy, UK.