Credit Guarantee and Fiscal Costs

• Published date: 01 August 2024
• Author: HUIXIN BI,YONGQUAN CAO,WEI DONG
• Date: 01 August 2024
• DOI: http://doi.org/10.1111/jmcb.13012

DOI: 10.1111/jmcb.13012

HUIXIN BI

YONGQUAN CAO

WEI DONG

Credit Guarantee and Fiscal Costs

This paper studies the effectiveness of government-backed credit guaran-

tees to the infrastructure sector. We propose a two-sector model with -

nancial intermediary frictions so that infrastructure producers rely on bank

loans to nance production. Governments can intervene in the credit market

by providing a partial guarantee. We nd that a credit guarantee increases

infrastructure production, leading to a high scal multiplier in the longer

run. In the near term, however,higher infrastructure-sector wages crowd out

private-sector labor supply.Importantly, the higher leverage associated with

credit expansion raises nonperforming loans, and this channel is particularly

pronounced if the government-backed credit guarantees linger.

JEL codes: E62, E44

Keywords: scal multiplier, credit guarantee, bailout policy, nancial

friction

I    2008 Global Financial Crisis, many

emerging market economies sought to boost their economies through government-

backed credit expansion targeting to specic sectors, in particular related to infras-

tructure. The persistence associated with those credit expansion, however, varied

across countries, leading to different patterns in nonperforming bank loans. Credit

guarantees that lingered well beyond the crisis led to a surge in nonperforming bank

loans in countries like India and China, while short-lived credit expansions were as-

sociated with stable nonperforming loans (NPLs) in countries like South Korea.

Wethank Tatiana Damjanovic, Stefano Gnocchi, Lena Suchanek, and seminar participants at the Bank

of Canada, Indiana University, Stockholm China Economic Research Institute, and 2018 International

Conference on Macroeconomic Analysis and International Finance for helpful comments. The views ex-

pressed in this paper are those of the authors and do not represent those of the Bank of Canada, the IMF,its

Executive Board or IMF management, the Federal Reserve Bank of Kansas City,or the Federal Reserve

System.

H B is a Research and Policy Ofcer, Research Department, Federal Reserve Bank of Kansas

City (E-mail: huixin.bi@kc.frb.org). Y Cis an Economist, African Department, International

Monetary Fund (E-mail: ycao2@imf.org). WD is a Research Advisor,Canadian Economic Analysis

Department, Bank of Canada (E-mail: wdong@bank-banque-canada.ca).

Received December 18, 2020; and accepted in revised form August 30, 2022.

Journal of Money, Credit and Banking, Vol. 56, No. 5 (August 2024)

© 2023 The Ohio State University.

1204 :MONEY,CREDIT AND BANKING

Although the credit guarantee has been widely used by policymakers, the literature

has yet to explore the impact of the persistence of those programs on the economy,

which is at the heart of our paper. We propose a two-sector model with nancial inter-

mediary (FI) frictions to study the transmission channels of government-backed credit

expansion to infrastructure sector. We nd that credit guarantee increases infrastruc-

ture production, leading to a high scal multiplier in the longer run. In the near term,

however, higher wages in the infrastructure sector crowd out labor supply in the pri-

vate sector, dampening economic activities. Importantly, higher leverage associated

with credit expansion raises NPLs, and this channel is particularly pronounced if the

government-backed credit guarantee lingers for a long period of time.

Specically,our model consists of a private sector and an infrastructure sector, with

imperfect substitutability of labor supply in the two sectors. Infrastructure goods are

converted into public capital, which poses positive externality on the productivity of

private sector.Importantly, infrastructure producers face a working capital constraint

and borrow from banks to nance their operation costs before production takes place.

Those infrastructure projects, however, are risky and producers may default on their

loans if their idiosyncratic productivity turns out to be lower than the break-even

threshold. Financial intermediaries set the loan contract to balance the trade-off be-

tween loan returns and NPLs. The government credit guarantee program, which pro-

vides partial guarantee or bailout on bank loans to infrastructure producers, changes

the incentives for banks to lend and, therefore, the loan contract. For a given produc-

tivity break-even level, a more generous bailout policy makes banks willing to lend

more and accept higher leverage. In turn, the change in the loan contract motivates

infrastructure producers to accept a higher break-even threshold, raising the NPLs.

We nd that governmentcredit guarantee affects the macroeconomy through three

offsetting channels, and the macroeconomic impacts of government credit guarantee

change over time. Credit easing relaxes the working capital constraint for infrastruc-

ture rms, boosting infrastructure production. Higher infrastructure capital raises the

productivity of private sector overtime, through the positive externality channel, and

therefore boost economic growth. In the near term, however, higher wages in the in-

frastructure sector can crowd out labor supply to the private sector through the wage

spillover channel, dampening economic activities. Finally, a higher leverage associ-

ated with credit expansion increases the NPL ratio for infrastructure rms, raising

bailout costs for the government. In the short run, the wage spillover and the bailout

cost channels dominate, leading to a negative scal multiplier on impact. Over the

longer run, however, the positive externality channel prevails over the bailout cost

channel, as higher infrastructure capital makes private rms more productive and

raise scal multipliers in the median term.

We also nd that the effectiveness of government credit guarantee crucially de-

pends on the persistence of credit expansion. A transitory and well-targeted credit

expansion has a much higher scal multiplier than a government-backed credit guar-

antee that lingers for several years. Specically, a transitory credit expansion can

generate a scal multiplier close to 2 over the medium and long run, as positive exter-

nality associated with higher public investment dominates the relatively low bailout

HUIXIN BI, YONGQUANC AO,AND WEI DONG :1205

costs as well as the short-term crowding-out effects. A persistent credit expansion,

on the other hand, can have negative scal multipliers both on impact and over time.

Signicantly higher NPLs lead to elevated bailout costs, weighing on economic ac-

tivities.

In addition, we compare the scal multiplier of targeted credit easing with those

associated with other conventional scal instruments. In the near term, the credit

guarantee has a lower scal multiplier when compared to stimulus measures through

higher government consumption or investment. Over time, a well-targeted credit ex-

pansion becomes more effective in stimulating the economy. Also, the effectiveness

of credit guarantee also depends on scal nancing schemes, as scal multipliers are

notably lower if government has to collect distortionary taxes, instead of lump-sum

taxes, to nance spending.

Our paper is closely related to the literature of scal multipliers and public invest-

ment. Leeper, Walker, and Yang (2010) conduct a positive analysis of infrastructure

investment by modeling implementation delays associated with infrastructure spend-

ing and also differentiating scal nancing schemes. They nd that long implemen-

tation delays, as well as distortionary scal nancing, can signicantly dampen the

stimulative effects of infrastructure spending. Chang et al. (2021) study how scal

policy changes in the public sector spillover to private sector. Our paper extends this

literature by studying government-backed credit expansion that specically targets

the infrastructure sector.

In addition, our paper is related to the broad literature of nancial frictions. In

particular, we extend the framework in Bernanke, Gertler, and Gilchrist (1999) to

incorporate potential government bailouts on nonperforming bank loans. Chang et al.

(2019) uses a similar framework to study optimal reserverequirement policy in China.

We differ from their paper by focusing on government-backed credit guarantee as

an unconventional scal instrument. Specically, the government uses this policy

tool to target the infrastructure sector, as adopted by many countries following the

2008 crisis. We highlight that the effectiveness of credit guarantee depends on the

persistence of credit policy and also its associated scal nancing schemes.

Finally, recent studies suggest that the stimulus impacts of the federal credit pro-

grams in the U.S. were likely to have been similar in magnitude to what was provided

by the American Recovery and ReinvestmentAct of 2009, but there is a lack of under-

standing of the mechanism, see Lucas (2016). Correia et al. (2021) show that credit

subsidies can overcome the zero lower bound constraint on interest rate policy. Our

paper adds to the discussion by exploring the transmission channels through which

credit policy affects economic activities.

The paper proceeds as follows. We start by describing the background of credit

expansions in India, China, and South Korea following the 2008 crisis. Section 3

develops the baseline model with government-backed credit expansion targeting the

infrastructure sector. Section 4 introduces the calibration and quanties the impact of

credit easing. Finally, Section 5 concludes.