Is Historical Cost Accounting a Panacea? Market Stress, Incentive Distortions, and Gains Trading

• DOI: http://doi.org/10.1111/jofi.12357
• Author: ANDREW ELLUL,CHOTIBHAK JOTIKASTHIRA,CHRISTIAN T. LUNDBLAD,YIHUI WANG
• Date: 01 December 2015
• Published date: 01 December 2015

THE JOURNAL OF FINANCE •VOL. LXX, NO. 6 •DECEMBER 2015

Is Historical Cost Accounting a Panacea? Market

Stress, Incentive Distortions, and Gains Trading

ANDREW ELLUL, CHOTIBHAK JOTIKASTHIRA, CHRISTIAN T. LUNDBLAD,

and YIHUI WANG∗

ABSTRACT

Accounting rules, through their interactions with capital regulations, affect financial

institutions’ trading behavior.The insurance industry provides a laboratory to explore

these interactions: life insurers have greater flexibility than property and casualty

insurers to hold speculative-grade assets at historical cost, and the degree to which

life insurers recognize market values differs across U.S. states. During the financial

crisis, insurers facing a lesser degree of market value recognition are less likely to sell

downgraded asset-backed securities. Toimprove their capital positions, these insurers

disproportionately resort to gains trading, selectively selling otherwise unrelated

bonds with high unrealized gains, transmitting shocks across markets.

THIS PAPER EXPLORES THE TRADING INCENTIVES of financial institutions induced by

the interaction between regulatory accounting rules and capital requirements

(Heaton, Lucas, and McDonald (2010)). The theoretical literature (e.g., Allen

∗Andrew Ellul is with Indiana University, CEPR, CSEF, and ECGI, Chotibhak Jotikasthira

and Christian T. Lundblad are at the University of North Carolina at Chapel Hill, and Yihui

Wang is at Fordham University. We thank Ken Singleton (the Editor) and two anonymous ref-

erees for many helpful suggestions. We are also grateful for comments received from Bo Becker,

Utpal Bhattacharya, Dirk Black, Mike Burkart, Charles Calomiris, Mark Carey, Kathleen Han-

ley, Scott Harrington, Cam Harvey, Jean Helwege, Alan Huang, Gur Huberman, Tim Jenkinson,

Sreenivas Kamma, Andrew Karolyi, Mozaffar Khan, Peter Kondor, WayneLandsman, Mark Lang,

Luc Laeven, Martin le Roux, Christian Leuz, Dong Lou, Ed Maydew, Colin Meyer, Greg Niehaus,

Marco Pagano, ˇ

Luboˇ

sP

´

astor, Adriano Rampini, Richard Rosen, Chester Spatt, Dragon Tang,Dim-

itri Vayanos, Vish Viswanathan, James Wahlen, Tzachi Zach; participants at the Adam Smith

Workshop in Corporate Finance (2012), American Finance Association Annual Meeting (2014),

City University of Hong Kong Conference, China International Conference in Finance (2012),

FIRS Annual Meeting (2012), NBER Workshop on Credit Rating Agencies (2013), Seventh NY

Fed/ NYU Stern Conference on Financial Intermediation (2012), SIFR Workshop on the Role of the

Insurance Industry (2014), and Tel Aviv University Finance Conference (2012); and seminar par-

ticipants at Chinese University of Hong Kong, Duke University, Einaudi Institute for Economics

and Finance, Imperial College London, Indiana University,International Monetary Fund, London

Business School, London School of Economics, Universidade Catolica Portuguesa, University of

North Carolina (Accounting), University of Oxford, University of South Carolina, University of

Toulouse, University of Virginia, and University of Warwick. We are especially grateful to Robert

Hartwig of the Insurance Information Institute for detailed discussions and Victoria Ivashina for

state-level regulator contacts. This research was conducted, in part, while Lundblad was visiting

the Einaudi Institute for Economics and Finance in Rome, Italy. The authors do not have any

conflicts of interest, as identified in the Disclosure Policy.

DOI: 10.1111/jofi.12357

2489

2490 The Journal of Finance R

and Carletti (2008), Plantin, Sapra, and Shin (2008), and Sapra (2008)) argues

that mark-to-market (MTM), or fair value, accounting leads to the forced sell-

ing of assets by financial institutions during times of market stress, resulting

in a downward liquidity and price spiral and potential contagion effects for

other markets,1whereas historical cost accounting (HCA) may avoid fire sales

and contagion effects. In contrast to this view on HCA, we provide empirical

evidence showing that, when interacting with regulatory capital requirements,

HCA induces incentives to “gains trade” where, in order to shore up capital,

an institution selectively sells otherwise unrelated assets with high unrealized

gains.2

The role of MTM during the recent financial crisis has generated intense

debate. The accounting rules followed by financial institutions may appear to

simply be a measurement issue, which, in frictionless markets, is free of any

impact on economic fundamentals. However, when markets are illiquid and

trading frictions high, financial assets may temporarily trade at market prices

that are well below fundamental values (Duffie (2010)). In such an environ-

ment, write-downs and impairments associated with the deterioration of asset

prices will lead to an erosion of financial institutions’ capital base, potentially

forcing the liquidation of some assets. Allen and Carletti (2008) argue that,

in such an environment, HCA can help avoid fire sales. In a similar vein,

Plantin, Sapra, and Shin (2008) argue that MTM generates excessive volatil-

ity in prices, degrading their information content and leading to suboptimal

decisions by financial institutions.

However, HCA may also lead to inefficiencies. Financial institutions using

HCA have incentives to engage in selective asset sales aimed at the early

realization of earnings (Laux and Leuz (2009)). Indeed, Plantin, Sapra, and

Shin (2008) note that HCA is not immune to these inefficiencies in normal

times when asset prices are high. In this paper, we focus on the implications

of this incentive for gains trading and its impact on financial institutions’

portfolios during times of market stress. We argue that it is precisely during

these episodes that financial institutions have the greatest need to realize gains

in order to improve capital positions.

Tofocus ideas, consider a setting in which financial institutions are regulated

by a risk-weighted capital adequacy metric. For example, insurance regulators

employ the Risk-Based Capital ratio (RBC ratio)—the ratio of statutory equity

capital to required capital.3A low RBC ratio indicates financial weakness. Now

1This view has received support from the banking industry as well. In a letter to the SEC in

September 2008, the American Bankers Association was of the opinion that, among several factors

that led to the financial crisis, “one factor that is recognized as having exacerbated these problems

is fair value accounting.”

2Bleck and Liu (2007) theoretically examine the economic consequences of MTM and HCA. They

show that HCA may distort management’s incentives, and, in some cases, may induce behavior

similar to “gains trading” when management tries to signal good project quality to the market. See

also Berger, Herring, and Szego (1995).

3This ratio is similar in spirit to various capital ratios used by bank regulators. For more details

on the RBC ratio as well as the analysis that follows, please refer to Section I.

Is Historical Cost Accounting a Panacea? 2491

consider an institution that has invested heavily in asset-backed securities

(ABS). During the crisis of 2007 to 2009, many ABS were severely downgraded

by rating agencies, putting downward pressure on the institution’s RBC ratio.

Broadly speaking, such downgrades are likely to increase the institution’s RBC

(the denominator of the RBC ratio), as it is a function of each asset’s credit

rating. ABS downgrades may also decrease the institution’s statutory equity

capital (the numerator of the RBC ratio) if the downgraded assets are marked

to market or impaired.4Given this pressure, the institution then faces a stark

decision: either sell the downgraded ABS to reduce the required RBC or retain

them and raise additional capital elsewhere.

Because the downgraded assets experience significant price declines, a cru-

cial determinant of the institution’s decision is whether the price declines are

(or will soon be) reflected in its statutory equity capital. This is where the

accounting treatment of these assets is likely to have a first-order effect on

trading and portfolio choices. If the downgraded asset is held under MTM, the

price decline would be automatically reflected in the balance sheet, and the loss

would directly reduce the institution’s statutory equity capital. From a purely

accounting perspective, the institution would be indifferent between keeping

the asset and selling it. However, from a regulatory capital perspective, selling

the downgraded asset has an important advantage, as swapping a speculative-

grade asset for either cash or an investment-grade asset immediately reduces

the required RBC and improves the RBC ratio. Taken together, selling the

downgraded asset is unambiguously beneficial if the asset is held at market

value.

The situation is very different if the downgraded asset is held under HCA,

as the decline in value would not be recognized in the balance sheet unless the

institution sold the asset. Selling the asset thus has two opposing effects on

the RBC ratio: (i) a positive effect from reducing the required RBC, and (ii) a

negative effect from recognizing the price decline in its statutory equity capital.

If the price decline were very large, as was the case for many downgraded ABS

in the recent crisis, the negative effect would likely dominate and selling the

asset would not be beneficial. To maintain a healthy RBC ratio, the institution

may need to resort to other measures. It is precisely in this situation that

the incentive for gains trading is increased: in order to shore up its capital

positions, the institution may selectively sell those assets held under HCA

with the largest unrealized gains, as doing so these gains are realized and flow

to its statutory equity capital.

In this paper, we use the insurance industry as a natural laboratory to in-

vestigate the above arguments, in particular, whether financial institutions

using HCA for troubled assets, compared to those using MTM, are less likely

to directly sell the troubled assets and more likely to gains trade. Under the

4In Section I.A, we demonstrate that, under the NAIC model regulation, downgrades of ABS

from investment to speculative grade reduce the RBC ratios of virtually all insurers holding these

assets. However, the precise impact depends on the severity of the downgrades and the associated

price declines, as well as the accounting treatment used to book the downgraded assets.