Why are Indian firms five times less productive than American ones? One reason is land.

|

This page in:

How important is land as a factor of production in India? In a recent paper (The misallocation of land and other factors of production in India," Policy Research Working Paper Series 7221), we focused on its role, and examined how the misallocation of land has impacted the country’s manufacturing sector.
 
India provides a rich ground for examining fundamental questions like how important addressing land market distortions is for growth. Firms in India differ enormously compared to USA. The productivity of a US firm at the top decile is typically twice as high as that of a firm at the bottom decile in a typical manufacturing industry. This increases to five times more in India. This suggests considerable factor market misallocation within India. A similar narrative can be found in other countries in Africa and East Asia.
 
We computed misallocation indices for different factors of production (labor, land and buildings, other fixed assets) as well as for output for the manufacturing sector. This was computed at the sub-national district level. This approach enables us to assess the effects of ‘frictions’ on economic development, and in particular focus our attention on how the misallocation of land has impacted output per worker.
 
India’s spatial disparities in output per worker show considerable variation across districts. The spatial disparity in productivity is huge and differences in factor misallocation within India are larger than the differences found in many other countries. Districts with higher output per worker are largely concentrated in Gujarat, Rajasthan, Haryana, Punjab, and Maharashtra. Low output per worker is more prevalent in the eastern and poorer lagging regions of India (Bihar, UP, and Orissa).  
 
While more productive enterprises manage to produce more than less efficient enterprises, the allocations of some factors of production are barely better than random. Indeed, there are many districts in India where factor allocation is worse than random. Two more subtle patterns also emerge. First, the misallocation of land and labour is worse than the misallocation of buildings (when it can be separated from land) and other fixed assets. Second, it is unfortunate that there appears to be a mild trend towards a worsening output and factor misallocation over time.
 
What was the impact of other factor market distortions? When we separate capital and employment, we estimate coefficients of roughly the same magnitude for both factors to explain the misallocation of output (or value added). When we consider land and buildings separately from other fixed assets, the coefficient on land and building is always much larger than that on other fixed assets. This is a striking result.  In one simple regression, one standard deviation in land and buildings misallocation is associated with 0.62 standard deviation of value added misallocation. The corresponding figure for employment is only 0.40. Even though land and building account for a small fraction of final output and value added, they seem to play a disproportionate role in explaining the misallocation of final output. Misallocation of capital appears to account for very little of the misallocation of final output. Overall, this finding is consistent with the notion that land is the most distorted factor market and the least flexible factor of production. It is likely that huge land misallocation is also breeding the misallocation in other factors of production.  
 
An important takeaway of our finding is that a standard deviation increase in the misallocation of all factors is associated with a 27% decrease in output per worker and most of this decline originates from the misallocation of land and buildings. While the gains from expanding supply, and particularly the supply of land are large, the gains from a better land allocation is even larger. When we correlate the misallocation of land and buildings at the district level with the misallocation of output and with the output per worker, land and buildings misallocation appears to be at the root of much of the misallocation of output and it accounts for a large share of the observed differences in output per worker. More precisely, across districts, a standard deviation in the misallocation of land and buildings accounts for about 20% difference in output per worker.
 
This line of work could be further extended to better understand what should be the priorities for policy action. Knowing about the within/between breakdown of misallocation is important since the policy tools to target both types of misallocation are likely to differ. For instance, further integration of financial markets within India or a reduction in the cost of migration between areas will likely affect misallocation (of capital and labour) across places first and foremost. Other policies such as a relaxation of the strict zoning policies in some cities, on the other hand, are expected to improve the efficiency of factor allocation in those areas where these regulations are the most binding. There is also a need to examine how factor market distortions are interacting with other emerging trends in urbanization, transport, and technology, given that the large manufacturing firms are leaving the mega cities and moving into intermediate size cities in search of lower land costs. While these relocations may be rational and justified from the perspective of the firm, they may worsen the allocation of factors in the country.

Alternatively, the worsening of factor allocation within district may be offset by an improvement of factor allocation across districts and states.  Furthermore, the within/between decomposition of misallocation could be applied to the formal and informal sectors and how these are being impacted in the tradable and non-tradable sectors, and the gender component of the enterprises. While one expects some misallocation among male-owned and among female-owned enterprises, assessing how much aggregate misallocation is accounted for by female-owned enterprises not having access to factor inputs is of first-order importance to assess the economic consequences of gender discrimination in India. Finally, firms face constraints on their external finance that depend on the amount of land and building that they own, and so tracking how financial constraints and what firms own in terms of real assets affect how firms operate and produce is also a first-order question.
 
Emerging Lessons

So what are some of the emerging lessons on land misallocation and growth? Policy makers should continue to give importance to factor accumulation, but improved factor allocation, particularly land, can be an even bigger driver of growth.  The misallocation of factor inputs hampers firm performance, and land and building distortions appear to be especially important. A one standard deviation decrease in the misallocation of land and buildings can bring about a 25% increase in output per worker in India. This is equivalent to a six-fold increase in the land supply for manufacturing establishments in these districts. More attention should also be given towards reducing huge spatial variation in allocation efficiency across districts and states. This spatial variation within India is of the same magnitude as the variation in allocation efficiency across countries of the world. Even though land and building account for a small fraction of final output and value added, they play a disproportionate role in explaining the misallocation of final output. Land is the least flexible factor of production and its misallocation also likely breeds the misallocations of other factors and output.
Topics

Authors

Gilles Duranton

Professor of Real Estate, The Wharton School

Arti Grover

Consultant, World Bank

Join the Conversation

Mariko Peters
October 13, 2015

Dear Ejaz Ghani, thank you for the interesting blog. Could you please explain: what is meant exactly by 'misallocation of land'? In what way is the misallocation manifested, how was misallocation determined?
Thank you,
Mariko Peters (Justice, WBG Nairobi)

Ejaz Ghani
October 13, 2015

Dear Mariko, 
There are many studies arguing the importance of misallocation. The consequences
of misallocation are usually inferred indirectly by asking a model what would be the aggregate consequences of reduced misallocation. That is, extant claims about the importance of misallocation rely on measures of aggregate misallocation and computations of counterfactuals from particular models. In contrast, we provide direct evidence about the importance of misallocation by investigating empirically the link between factor misallocation across establishments and output per worker, using subnational regions in cross-section and in panel.
 
Increasing the productivity of factors of production, fostering their accumulation, and reducing their misallocation can only be viewed as proximate causes of economic growth and development. Importantly, as part of our metric validation, we begin to make connections between policies and factor misallocation. Our findings suggest that factor misallocation is not exogenously determined but is instead affected by policies. More generally, we think of our results as emphasizing the importance of “frictions” as key determinant of the efficiency of factor allocation and, in turn, prosperity. Policies can be a source of friction. Better policies can also reduce frictions and improve allocation. Our emphasis on frictions differs from many models in the literature that rely on idiosyncratic distortions as the root cause of misallocation.
 
We highlight the uniquely important role played by land and buildings in misallocation. We attribute this to the fact that choosing a location is a decision that conditions many others and cannot easily be changed, especially in an environment with poorly functioning land markets. More productive establishments will have a tough time buying more machines or employing more workers if they have no room to accommodate them. Land may also be a uniquely important asset for establishments that seek to expand since it can be used as collateral for external financing. While we do not take a stance regarding the exact mechanism through which land and buildings affect factor misallocation, our results highlight that land used for nonagricultural production may play a more important role than hitherto thought. Better land policies can make land more broadly available and reduce the frictions associated with land transactions.
We are nonetheless aware that the elimination of frictions on the land market would require more than better land use regulations and a more efficient taxation of properties. Better-functioning land markets also require clearly defined property rights, a reliable land registry, and a number of other institutional improvements. This is a considerable challenge in a country like India where property rights for land and buildings are poorly defined and often conflict with tenancy rights.
 
Our analysis proceeds in five steps. The first is to estimate the productivity of
all establishments in the data and factor shares for all industries. Establishment productivity is a necessary input to measure misallocation, and we consider several approaches to estimating productivity. Our second step is to compute misallocation indices for output and for each factor of production at the district level. The main difficulty here is that misallocation is most meaningfully computed at the industry level since industries differ in their factor intensity. Measures of misallocation at the district-industry level must thus be aggregated across industries to obtain a district-level measure. After distilling some stylized facts about misallocation across districts, our third step validates these metrics for studying land and building access by showing a connection of them to unanticipated local policy reforms that affect property markets. In the fourth step, we quantify the relationship between various forms of misallocation. This step allows us to tease out the unique role played by the misallocation of land and buildings among Indian establishments. This misallocation of land and buildings is at the root of much of the misallocation of output. The last step examines the effects of factor misallocation on establishment productivity and output per worker. This analysis affords statements about which forms of misallocation matters more in Indian districts. Again, land and buildings appear especially important given their relatively small cost share.