Loss Shocks in Export Credit Insurance Markets: Evidence From a Global Insurance Group

• Date: 01 March 2019
• Author: Koen J. M. van der Veer
• DOI: http://doi.org/10.1111/jori.12197
• Published date: 01 March 2019

©2017 The Journal of Risk and Insurance. Vol.86, No. 1, 73–102 (2019).

DOI: 10.1111/jori.12197

Loss Shocks in Export Credit Insurance Markets:

Evidence From a Global Insurance Group

Koen J. M. van der Veer

Abstract

Private export credit insurance—covering the risk of nonpayment—plays an

important role in facilitating international trade, especially within Europe.

Due to lack of data, however,little is known about the influence of loss shocks

on export credit insurance markets. This article studies the effect of claims

on the availability and premium of export credit insurance, using unique bi-

lateral country-level data covering worldwide insurance underwriting from

1992 to 2006 by a leading trade credit insurance group. Applying fixed ef-

fects models at the country subsidiary level, I find that a doubling of the

claims ratio on insured exports between a pair of countries results, on av-

erage, in a decline in the subsidiary’s share of bilateral exports insured by

about 11 percent and rise in premium level by about 4 percent.These claims

effects increase when the insurer makes a loss and rise with the size of the

loss. Importantly, evidence shows that an extreme loss shock in one market

also increases the claims sensitivity of insurance coverage on exports to other

markets, suggesting a role for capital constraints. Overall, these results help

our understanding of potential trade finance constraints in times of crisis,

such as during the 2008–2009 global trade collapse.

Introduction

For decades, developments in trade credit insurance markets remained largely unno-

ticed to policy makers and the wider public. For example, the structural change within

the European Union (EU)—where public export credit agencies were restricted and

the market for private short-term trade credit insurance developed into a multibillion

Koen J. M. van der Veer is at the De Nederlandsche Bank, The Netherlands, P.O. Box 98, 1000

AB, Amsterdam. The author can be contacted via e-mail: k.j.m.van.der.veer@dnb.nl. I thank

Paul Becue, Jan Willem van den End, Marco Hoeberichts, Neeltje van Horen, Eelke de Jong,

Andrew Rose, Job Swank, Marco Verwoerd, seminar participants at De Nederlandsche Bank,

and two anonymous referees for useful comments. Henk van Kerkhoff providedexcellent data

assistance. I bear full responsibility for any remaining errors. An earlier DNB Working Paper

version of this article was circulated under the title “Loss Shocks and the Quantity and Price of

Export Credit Insurance: Evidence From a Global Insurer.”The views expressed in this article

are those of the author and do not necessarily represent those of De Nederlandsche Bank.

73

74 The Journal of Risk and Insurance

business—took place silently since the late 1990s.1As a result of this change, export

credit insurance covering the risk of nonpayment on short-term trade transactions

within the EU is now generally provided by private insurers (Morel, 2011). World-

wide, private trade credit insurers—“niche” players within the broader market for

trade finance—covered EUR 2.2 trillion of international and domestic trade in 2014

(ICISA, 2016).

Developments in export credit insurance markets did make it to the global policy

agenda during the 2008–2009 global trade collapse. In 2008, the trade credit insurance

industry faced a rapidly deteriorating risk environment and an almost doubling of

the claims ratio (claims over premium income) on insured domestic and international

trade to 85 percent, compared to an average of 43 percent in the 5-year period before

the crisis (ICISA, 2016). As a reaction, trade credit insurers cut back on their insurance

supply and raised premium rates. On the demand side, there were opposing influ-

ences. The decline in global demand reduced world trade and thereby the need for

export credit insurance, while at the same time, therewas high demand from exporters

for credit insurance protection (Morel, 2011). These circumstances fueled the debate

on a possible gap in the supply of short-term export credit insurance, which led to

the G-20 call on public export credit agencies to fill the perceived gap.2Yet, the extent

of the “hard market”—an insurance market characterized by a limited availability of

insurance and higher premium rates—in short-term export credit insurance was dif-

ficult to assess. Real-time figures on the quantity and premium level of private export

credit insurance are publicly unavailable and research on the conduct of private trade

credit insurers did not exist.3

This article is the first to study the effect of claims incurred by a private trade credit

insurer on its quantity and premium of export credit insurance. Using a unique bi-

lateral country-level data set on insurance underwriting by one of the “Big Three”

global trade credit insurance groups, I first estimate the average (benchmark) effect

of a temporary increase in claims on the quantity and premium of export credit in-

surance at the country subsidiary level. Holding other things constant, on average, a

1Since 1998, EU legislation restricted public export credit agencies from providing guarantees

covering export credit risks with a maturity of less than 2 years to the “marketable risk coun-

tries,” that is, the countries within the EU and most other OECD countries (see European

Commission, 1997).

2Subsequently, 14 governments within the EU set up various state aid schemes to support the

market for short-term export credit insurance (for an overview, see van der Veer,2011).

3The Bern Union (BU) provides the most extensive publicly available data on short-term trade

credit insurance. A few limitations of these country-level data, however,make it difficult to as-

sess developments in the market for private short-term export credit insurance. First, the data

combine short-term export credit insurance provided by private insurers and public export

credit agencies (ECAs). This is likely to mask changes in private insurance cover, especially

during a crisis when ECAs aim to support the market. Second, the BU collects data by destina-

tion country only; thus, no inference can be made on home country developments in insurance

coverage. Last but not least, the BU quarterly figures report “credit limits,” which refer to the

amount of total potential exposure by insurers and do not measurethe amount of actual export

credit insured.