Testing Household‐Specific Explanations for the Inverse Productivity Relationship

Abstract

The inverse relationship between land productivity and farm size is an old and puzzling empirical regularity. Most explanations for this relationship rely on market imperfections that jointly determine the farm size and the household's shadow price of some productive inputs. We use plot-level data from the ICRISAT/VLS to assess whether these household-specific theories can explain the puzzle. The data exhibit plots of different sizes being simultaneously cropped by the same household. The inverse relationship is shown to hold true with the same magnitude across the plots of each household, thus cross-household heterogeneity does not suffice to explain the puzzle.

Looking at Pro-Poor Growth from an Agricultural Perspective

“…A particularly well-known argument suggests that family and hired labor are actually only imperfect substitutes and that labor productivity of hired workers on large farms is positively associated with the level of supervision by the landowner (Feder 1985;Frisvold 1994). Hence, large landowners will regularly exhibit a higher optimal land-to-labor ratio than smallholders and -given imperfections on the land market -the IFSP relationship will emerge (Assunção and Braido 2007). 7 The underlying assumptions are that if better soil quality actually leads to higher output and if soil quality has a negative correlation with farm size, then the IFSP may be detected if the analysis does not adequately account for differentials in soil quality (Barrett et al 2010).…”

Small is Beautiful: Evidence of an Inverse Relationship Between Farm Size and Yield in Turkey

“…It is often argued that the inverse size-yield relationship is a statistical artifact due to omitted variables (Bhalla and Roy 1988;Benjamin 1995;Assuncao and Braido 2004).…”

“…Hence, indirect methods of accounting for land quality must be applied. These include relying on geographical disaggregation (Rudra 1974;Sen 1981;Carter 1984;Bhalla and Roy 1988), using price of land or share of irrigation as a proxy for land quality (Khusro 1973;Berry and Cline 1979), using village or plot fixed effects (Carter 1984;Heltberg 1998;Assuncao and Braido 2004), and employing instrumental variables to proxy for land quality (Benjamin 1995).…”

“…By using either the exact or some modified version of the classical equation (1) (Bhalla and Roy 1988;Benjamin 1995;Benjamin and Brandt 2002;Lamb 2003;Assuncao and Braido 2004;Srinivasan 1972;Chayanov 1966) and the factor market imperfection hypothesis (Sen 1962(Sen , 1966Mazumdar 1965;Sen 1981;Cornia 1985;Eswaran and Kotwal 1986;Heltberg 1998;Benjamin and Brandt 2002).…”

Land Measurement Bias:

“…The inverse land size-productivity relationship has been well documented by Carter (1984); Barrett (1996); Assunação and Braido (2007); Barrett et al (2010);Carletto, Savastano, and Zezza (2013);and Carletto, Gourlay, and Winters (2015), among others. Using self-reported plot size and Google measurements, we estimate equation (2) where ����� �� is the natural logarithm of total output value of � �� on plot �, in the remotely sensed area mesh m, divided by plot size (���� �� ), where the observations are stacked such that each plot has an observation using GPS, SR, and Google measurement techniques and quartiles by land measurement method interactions are included.…”