External Financing in the Life Insurance Industry: Evidence From the Financial Crisis

• DOI: http://doi.org/10.1111/jori.12042
• Date: 01 September 2014
• Published date: 01 September 2014

EXTERNAL FINANCING IN THE LIFE INSURANCE INDUSTRY:

EVIDENCE FROM THE FINANCIAL CRISIS

Thomas R. Berry-Sto

¨lzle

Gregory P. Nini

Sabine Wende

ABSTRACT

The financial crisis and subsequent recession generated sizable operating

losses for life insurance companies, yet the consequences were far less

significant than for other financial intermediaries. The ability to quickly

generate new capital through external issuance and dividend reductions let

life insurers maintain healthy levels of equity capital. We use this experience

to examine the causes and consequences of external capital issuance by U.S.

life insurance companies. We show that, in general, new capital is issued

both to support the growth of new business and to replace capital depleted

by operating losses. This second channel is particularly important during

macroeconomic recessions. Notably, we do not find any evidence that

insurers had difficulty generating new capital, unlike other financial service

providers that required large amounts of public support. For life insurers,

what changed following the financial crisis was the demand to raise external

capital, but the supply of external capital appears to have remained constant.

INTRODUCTION

On the heels of the financial crisis and subsequent recession, roughly 30 percent

of U.S. life insurance companies raised new capital during 2008 and 2009, about

Thomas R. Berry-Sto

¨lzle is at the Terry College of Business, University of Georgia, 206 Brooks

Hall, Athens, GA 30602. Berry-Sto

¨lzle can be contacted via e-mail: trbs@uga.edu. Gregory P.

Nini is at the LeBow College of Business, Gerri C. LeBow Hall 1129, 3141 Chestnut Street,

Philadelphia, PA 19104. Sabine Wende is at the Faculty of Management, Economics and Social

Sciences, University of Cologne, Albertus-Magnus-Platz, 50923 Cologne, Germany. The

authors would like to thank participants of the 2013 AEA Annual Meeting, 2012 NBER

Universities Research Conference, the 2012 Risk Theory Society Seminar, the 2012 ARIA

Annual Meeting, the 2012 FMA Annual Meeting, and the Convergence, Interconnectedness,

and Crisis: Insurance and Banking Conference at Temple University for helpful comments and

suggestions. All remaining errors are our own. Thomas R. Berry-Sto

¨lzle gratefully acknowl-

edges financial support from a Terry-Sanford research grant. Sabine Wende gratefully

acknowledges financial support from the 2nd Female Professor Program of the University of

Cologne and travel support from the German Academic Exchange Service.

© The Journal of Risk and Insurance, 2014, Vol. 81, No. 3, 529–562

DOI: 10.1111/jori.12042

529

double the frequency during the prior 5 years. New issuance was concentrated in

insurers specializing in annuities, who suffered particularly large drops in

profitability due to their exposure to equity markets through variable annuity

products.

1

The large influx of new capital helped life insurers maintain capitalization

levels close to their historical levels, despite vastly reduced retained earnings.

In this research, we examine the causes and consequences of external capital issuance

by U.S. life insurance companies from 1999 through 2010, paying special attention to

the two recessions that occurred during this period. We show that capital issuance is

driven by two distinct forces: (1) the need to fund additional growth that cannot be

supported by existing capital and retained earnings, and (2) the need to replace

capital that has been depleted by losses from existing business. In regression results,

we show that new issuance is concentrated in firms with significant growth in

premiums written, negative net income, and low levels of capital. Importantly, there

is no evidence that the relationship between depleted capital and equity issuance is

different during recessionary periods, including 2008 and 2009. What changed

following the financial crisis was the need to raise external capital, but the ability to

tap external capital markets appears to have remained constant.

We show these results using life insurers’ financial statement data provided by A.M.

Best Company. The data are based on insurance companies’ regulatory filings with

the National Association of Insurance Commissioners (NAIC); A.M. Best organizes,

standardizes, and consolidates the financial statements of individual subsidiaries to

the group level. Our final data include insurance groups and single unaffiliated

insurance companies. Using consolidated financial statement data ensures that our

measures of external capital issuance are not confounded by internal capital transfers

within groups.

2

We construct indicators of capital issuance for new paid-in capital

and surplus notes based on insurer balance sheets and summaries of operations. New

paid-in capital includes common equity and preferred equity, and surplus notes

refers to a subordinated-debt instrument that is treated like capital for regulatory

purposes. Using logit regression models, we identify the partial correlations between

capital issuance and insurer-level characteristics that proxy for growth opportunities

and capital shocks. We show that firms with more growth options are more likely to

issue new capital, firms with less capital are more likely to issue new capital, and firms

with lower earnings are more likely to issue new capital.

We also examine the correlations between insurer characteristics and the decision to

cut dividends to stockholders, which is an alternative means to boost capital levels.

We find that firms cut dividends in response to low capital but do not adjust

dividends to support growth opportunities. During 2009, a large number of life

insurers cut dividend payments, which was a second means for them to boost

capitalization following the large shock in 2008.

1

These products often offered a guaranteed minimum rate of return on the investment

component of the annuity, which insurers did not completely hedge.

2

There is a substantial body of literature on capital transfers within conglomerates (e.g.,

Stein, 1997; Campello, 2002). For the property–liability insurance industry, Powell, Sommer,

and Eckles (2008) document the existence of such active internal capital markets.

530 THE JOURNAL OF RISK AND INSURANCE

In all of our analysis, we compare three distinct time periods: 2001–2002, 2008–2009,

and all other years. This approach allows us to examine the stability of the estimated

relationships across very different macroeconomic periods. In our most basic

analysis, we estimate the relationship between capital issuance and insurer

characteristics without accounting for calendar year. We then show that the large

increase in capital issuance during 2008–2009 is in line with what was expected based

solely on the changes in insurer characteristics, particularly the fall in earnings during

2008. Importantly, there is no evidence that actual issuance was below what would

have been predicted. More formally, we cannot reject the hypothesis that the

estimated relationships are stable across the three periods. Changes in capital

management across time reflect changes in insurer characteristics; given the large

shock to insurer earnings in 2008–2009, it is not surprising that a large number of

insurers raised new equity during this period.

We also explore the choice between surplus notes and other capital. Conditional on

issuing some new capital, we regress an indicator of surplus note issuance on a set of

insurer-level characteristics. We show that mutual insurers are more likely to issue

surplus notes and that insurers with lower capital levels are more likely to issue

surplus notes. Surplus notes offer a novel mechanism for mutual insurers to quickly

raise capital.

Finally, we examine the consequences of new capital issuance. Our results show that

firms issuing new capital tend to experience growth in premiums and also build their

capital levels. Issuers of new capital also are less likely to pay dividends, confirming

that insurers use both sources of capital to replenish lost capital. Importantly, these

relationships are no weaker during the 2008–2009 recession, suggesting that access to

external capital was particularly valuable during this period. Indeed, in many cases

we find that the effects are stronger during 2008–2009 than in the prior recession and

in nonrecessionary periods.

Our results have a direct implication for government policy that was enacted by

the U.S. Treasury following the crisis. In May 2009, the Treasury’s Troubled Asset

Relief Program (TARP) was extended to life insurers, providing them with access to

government-provided capital.

3

Two of the largest life insurers—The Hartford and

Lincoln National—chose to solicit and receive TARP funds.

4,5

The typical justification

3

The law permitted TARP to provide funds to life insurers with bank holding company status

or a thrift subsidiary.

4

The Hartford received $3.4 billion, and Lincoln National received $950 million (source: SNL

Financial’s TARP Participant List). Four more insurers—Allstate, Ameriprise Financial,

Principal Financial, and Prudential Financial—applied for and were authorized to receive

TARP funds but ultimately decided against receiving the funds. Glenworth Financial applied

for TARP funding but the application was rejected. See Harrington (2009) and Grace (2011) for

more on TARP and life insurers.

5

The federal government bailout of American International Group (AIG) was also partially

financed with TARP funds. However, as Harrington (2009) documents in his detailed case

study, AIG had a unique structure including a subsidiary that wrote credit default swaps

(CDS). The CDS portfolio of this London-based subsidiary created the large losses that

EXTERNAL FINANCING IN THE LIFE INSURANCE INDUSTRY 531