The Impact of Incentives and Communication Costs on Information Production and Use: Evidence from Bank Lending

• Author: PHILIP E. STRAHAN,JUN (QJ) QIAN,ZHISHU YANG
• Published date: 01 August 2015
• Date: 01 August 2015
• DOI: http://doi.org/10.1111/jofi.12251

THE JOURNAL OF FINANCE •VOL. LXX, NO. 4 •AUGUST 2015

The Impact of Incentives and Communication

Costs on Information Production and Use:

Evidence from Bank Lending

JUN (QJ) QIAN, PHILIP E. STRAHAN, and ZHISHU YANG∗

ABSTRACT

In 2002 and 2003, many Chinese banks implemented reforms that delegated authority

to individual loan officers. The change followed China’s entrance into the WTO and

offers a plausibly exogenous shock to loan officer incentives to produce information.

We find that the bank’s internal risk rating becomes a stronger predictor of loan

interest rates and ex post outcomes after reform. When the loan officer and the branch

president who approves the loan work together longer, the rating also becomes more

strongly related to loan prices and outcomes. Our results highlight how incentives

and communication costs affect information production and use.

HIGH-QUALITY INFORMATION is essential for successful business transactions. A

growing literature emphasizes how both individual incentives and the cost

of communication to decision-making authorities affect the production and

use of information. In this paper, we study how banks use information to set

loan interest rates, how that information forecasts future outcomes (default),

and how the quality (predictive power) of information production varies with

incentives and communication costs.

We use data from China, which historically has been dominated by large

and inefficient state-owned banks that rely on centralized decision-making pro-

cesses. Following China’s entrance into the WorldTrade Organization (WTO) in

∗Qian is at the Shanghai Advanced Institute of Finance, Shanghai Jiaotong University, and

is affiliated with the China Academy of Financial Research and Wharton Financial Institutions

Center. Strahan is at the Carroll School of Management, Boston College, and is affiliated with

the Wharton Financial Institutions Center and NBER. Yang is at the School of Economics and

Management, Tsinghua University. We appreciate helpful comments from Bruno Biais (Editor),

an Associate Editor, two anonymous referees, ViralAcharya, Daniel Carvalho, Andrew Hertzberg,

Olivier De Jonghe, Jongsub Lee, Chen Lin, Nadya Malenko, Daniel Paravisini, Amir Sufi, Weining

Zhang, and seminar/conference participants at Boston College, Cheung Kong Graduate School of

Business, Federal Reserve Banks of New York and Philadelphia, Shanghai Advanced Institute of

Finance, University of Florida, Bank Competitiveness Conference at Universit`

a Bocconi, China In-

ternational Conference in Finance (Wuhan), Financial Intermediation Research Society meetings

(Sydney), Summer Institute of Finance (Kunming), NBER China Workshop, and Utah Winter Fi-

nance Conference. We gratefully acknowledge research assistance from Shiyang Huang, Lei Kong,

and Wei Xiang, and financial support from Boston College and National Science Foundation of

China (Project numbers 70871071, 71272024, and 71232003). The authors are responsible for all

remaining errors.

DOI: 10.1111/jofi.12251

1457

1458 The Journal of Finance R

December 2001, however, many banks implemented a series of reforms during

the second half of 2002 and throughout 2003 focusing on decentralization—

shifting the responsibilities of making lending decisions from committees to

individuals.1These reforms strengthened incentives for loan officers to produce

and banks to use high-quality information, yet they are plausibly exogenous

from the perspective of loan officers because the reforms came from the highest

level due to external pressure.

We exploit proprietary loan-level data from a large, nationwide state-owned

bank that provides information on both interest rates and outcomes (full re-

payment on time, partial or late repayment, total loss). We test how incentives

to produce and use information affect, first, how banks set ex ante loan pricing

based on that information, and, second, how well that information forecasts

future loan performance. We then test how communication costs affect infor-

mation production and use, where costs are proxied by the time the information

producer (loan officer) and final decision maker (branch president) have worked

in the same branch.

Our sample covers borrowers located in more than 30 cities across China

over the 2000 to 2006 period. We treat the first half of 2002 and earlier as

the pre-reform period, and 2004 and later as the post-reform period. The key

information measure is the bank’s internally generated credit rating, which

measures the bank’s assessment of borrower risk. Before reform these ratings

were produced and approved by a group of loan officers from the bank’s loan

investigation unit; after reform, however, individual loan officers within the

unit become responsible for the ratings and can be held liable for bad loans

extended based on inaccurate or biased ratings.

In the first part of our analysis, we test the theoretical prediction that increas-

ing the authority and accountability of individual loan officers strengthens their

incentive to produce high-quality information, and such information is given a

more prominent role in the decision process (e.g., Aghion and Tirole (1997)). We

find that the bank places more weight on the credit rating in setting loan terms

after reform, conditional on borrower characteristics. Furthermore, a better

credit rating leads to a greater reduction in interest rates in the post-reform

period. These effects are stronger in coastal provinces, where the incentives

for loan officers to produce and branches to use high-quality information are

greatest. We next show that the information content embedded in the credit

rating and interest rate improves after reform—both become better predictors

of loan default. Thus, with better incentives the bank impounds better infor-

mation into loan interest rates, which in turn leads to interest rates’ greater

power to predict future default. These results are robust to the inclusion of

local banking sector competitiveness as well as the strength of past lending

relationships with the borrowers.

1The four largest state-owned banks have become publicly listed and traded on both domestic

and Hong Kong exchanges, with various government agencies retaining majority (equity) control.

These banks are currently among the largest banks in the world (source: Bloomberg). See, for

example, Allen et al. (2012) for more details.

Impact of Incentives and Communication Costs on Information 1459

In our second set of tests, we consider the effects of communication costs on

information production and use. Theoretical research (e.g., Crawford and Sobel

(1982), Bolton and Dewatripont (1994), Dessein (2002), Dewatripont and Tirole

(2005), Harris and Raviv (2005,2008)) shows that communication is costly

because it takes time and effort for an agent to absorb new information sent

by others and because agents may have (different) biases when sending and

interpreting information. In our setting we argue that, when the loan officer

and the head of the same bank branch (a key actor in loan approval) have

worked together longer, communication costs should be lower. Familiarity per

se does not guarantee better information—for example, a branch president/loan

officer pair who has worked together for an extended period may be more likely

to collude and favor questionable borrowers. In this regard, it is important

to note that we conduct these second set of tests on the post-reform data.

Because reform led to improved loan officer incentives, we expect a positive

incremental effect of time worked together between a loan officer/branch head

pair on information production as a result of lower communication costs.

As with reform, we find that the bank places greater weight on the credit

rating as the time worked together between a loan officer/branch head pair

increases. Moreover, we find that both the internal credit rating and the in-

terest rate better predict loan outcomes as the length of time between a pair

increases. We consider the possibility that these results reflect a spurious corre-

lation between loan officer quality and the length of the collaboration with the

president (e.g., low-quality officers may be more likely to be dismissed). The

result is robust, however, to controlling for both loan officer experience and

past performance, as well as for the branch president’s experience. In our view

adding these controls helps rule out endogenous matching as an explanation

for our findings, but we admit that we have no plausibly exogenous instrument

to fully resolve this concern. Nevertheless, the results support the idea that

lower communication costs improve both the quality of information production

and how that information is used.

Our paper contributes to and extends the literature on the role of informa-

tion in financial contracting. Despite ample theoretical work, there is limited

empirical validation of these theories. One difficulty has been a dearth of plau-

sibly exogenous variation across firms in incentive structures. An additional

obstacle has been the difficulty in finding measures of communication costs

that can be converted into quantitative variables, as well as measures of out-

comes to assess the quality of information produced. Our results, based on an

exogenous shock to the banking sector, detailed loan-level data including both

the terms and outcomes of loans, as well as job-related histories of loan offi-

cers and branch presidents, highlight the importance of incentive structures

and communication costs for the production, transmission, and use of informa-

tion. Better information, we find, expands the supply of credit and improves

(lending) outcomes.

A few recent empirical studies in banking explore information production

and usage, but are unable to exploit plausibly exogenous variation such as the