Tenure Security and Long‐term Investment on Tenanted Land: Evidence from Colonial Taiwan

Document

Cited in

Date	01 October 2019
DOI	http://doi.org/10.1111/1468-0106.12211
Author	Elliott Fan,Shu‐jen Yeh
Published date	01 October 2019

TENURE SECURITY AND LONG-TERM INVESTMENT ON

TENANTED LAND: EVIDENCE FROM COLONIAL

TAIWAN

ELLIOTT FAN*National Taiwan University

SHU-JEN YEH National Taiwan University

Abstract. We use farm diary data from Taiwan in the 1920s and 1930s to estimate the impacts of a

state-wide tenancy reform on tenants’investment in the farmland and production outcome. The

reform, commencing in 1922, enhanced the tenure security for the tenants by promoting written

contracts that extended tenure length from 1 year to 5–6 years, and by establishing government-

sponsored organizations for dispute settlements. Our estimations rely on a difference-in-difference

framework, where self-cultivated farms are used as the control group. We ﬁnd that the reform

encouraged tenants to make investments in the irrigation and drainage system of the farms, which

is expected to have long-lasting beneﬁts for production. The improved irrigation/drainage system

led to signiﬁcant efﬁciency gains for the farms. In contrast, the reform did not affect tenants’usage

of fertilizer much, which beneﬁts production for only a crop season or a year.

1. INTRODUCTION

Secure property rights are considered as an important premise for long-term

investment. Thus, a reduction of risk of losing the opportunity of harvesting

the returns in the future enhances the incentive to invest. The importance of

providing such an incentive for lasting economic growth has long been

recognized by previous researchers (North and Weingast, 1989; Besley, 1995).

The past two decades have witnessed a development of studies that provide

micro-data evidence to shed light on how individuals respond to an improve-

ment of property rights security. However, evidence from the existing studies

is dominantly focused on land ownership (Besley, 1995; Alston et al., 1996;

Besley and Burgess, 2000; Banerjee et al., 2002; Brandt et al., 2002; Jacoby

et al., 2002; Field, 2005 and Galiani and Schargrodsky, 2010); much less

evidence is provided to shed light on the contractual relationship between the

tenant and the landlord.1

In this paper, we estimate how tenants respond to a more secure tenure from

the perspective of agricultural investments and production efﬁciency. We exploit

an important historical reform of the tenancy system in Taiwan almost a century

ago: the introduction of Tenancy Improvement Organizations (TIO) in the

1920s. The reform endowed tenants with stronger tenure rights by making the

following two changes. First, it promoted formal written contracts between

the landlord and tenant in place of the traditionally prevailing oral contracts.

*Address for Correspondence: Department of Economics, National Taiwan University, Taipei,

Taiwan, ROC. E-mail: elliottfan@ntu.edu.tw.

1See Banerjee et al. (2002) as a rare example that explores the effects of a tenancy reform on agri-

cultural productivity in India.

Paciﬁc Economic Review,••:•• (2017)

doi: 10.1111/1468-0106.12211

In particular, in the design of the TIO written contracts, the contract length was

deliberately extended from 1 year, commonly adopted in paddy lands under oral

contracts, to 5 or 6 years. The extension of the contract length helped secure the

tenant’s usage right of the land, and, thus, incentivized the tenant to increase

long-term investment.

Second, the TIO contracts grant the authority of the TIO to settle contractual

disputes if they arise. When a private solution to a contract dispute cannot be

achieved, the unsettled case will be submitted to the TIO and processed by a

committee formed by government ofﬁcials and equal numbers of tenants and

landlords selected from the members of the TIO. The committee serves as an

impartial third party that offers a less costly channel for dispute settlement than

the ordinary judicial channels through court.

We exploit a very unique data set to implement our empirical analysis. The

data are drawn from diaries of rice farms during 1925–1927, and another round

of diaries (of different farms) during 1930–1931. These diaries detail prices and

quantities of the complete set of inputs and outputs associated with the farm’s

production, making it possible to apply advanced econometrics tools to conduct

estimations.2

To estimate the effects of the introduction of TIO, we design a difference-

in-difference strategy that focuses on comparing tenanted farms (as the

treatment group), which would potentially be affected by the introduction of

TIO, and self-cultivated farms (as the control group), which were unaffected

by the TIO reform. In addition, because TIO in different localities were

established in different years and months, we are able to measure the duration

of TIOs’operation in each locality and estimate the effects of the duration. This

provides an alternative perspective to examine the effects of TIO, and this

perspective is particularly important if the effects of TIO were not immediate,

but gradually intensiﬁed over time.

We examine three outcomes variables that represent various respects of a

farm’s performance. The three variables are: (1) proﬁt-cost ratio; (2) quality of

irrigation/drainage system; and (3) fertilizer usage density. The ﬁrst variable

has been widely used to measure the efﬁciency of the farm’s production and

management; the second and third variables are employed to indicate long-term

and short-term investments on a farm, respectively.

The main result from our empirical analysis suggests that the effect of the TIO

reform on improving the irrigation system is signiﬁcant. The proportion of farms

equipped with a top-quality irrigation/drainage system increased by 16–17

percentage points for tenanted farms after the establishment of TIO, relative

to the change for self-cultivated farms. However, we do not ﬁnd any signiﬁcant

effect of the establishment of TIO on fertilizer usage per acre. Combined, these

ﬁndings imply that that prolonging contract length and introducing a credible

third party for settling contract disputes enhanced long-term investment in

tenanted farms, but the impacts on short-term investment were only minimal.

2The 1930–1931 data lack information about the price of fertilizer, but this is not a threat to our

estimations. See more details in Section 3.

E. FAN AND S.-J. YEH2

Paciﬁc Economic Review

, 24: 4 (2019) pp. 570–587

doi:10.1111/1468-0106.12211