Choi 2013

This article was downloaded by: [Umeå University Library]
On: 03 October 2013, At: 13:24

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Asia-Pacific Journal of Accounting &

Economics
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/raae20
Asset revaluations: motives and choice

of items to revalue
a b b b
Tae Hee Choi , Jinhan Pae , Sunghwa Park & Younghyo Song
a
KDI School of Public Policy and Management , Seoul , Korea
b
Business School, Korea University , Seoul , Korea
Published online: 25 Sep 2012.
To cite this article: Tae Hee Choi , Jinhan Pae , Sunghwa Park & Younghyo Song (2013) Asset
revaluations: motives and choice of items to revalue, Asia-Pacific Journal of Accounting &
Economics, 20:2, 144-171, DOI: 10.1080/16081625.2012.719858
To link to this article: http://dx.doi.org/10.1080/16081625.2012.719858
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Asia-Pacific Journal of Accounting & Economics, 2013
Vol. 20, No. 2, 144–171, http://dx.doi.org/10.1080/16081625.2012.719858
Asset revaluations: motives and choice of items to revalue

Tae Hee Choia, Jinhan Paeb*, Sunghwa Parkb and Younghyo Songb
a
KDI School of Public Policy and Management, Seoul, Korea; bBusiness School, Korea
University, Seoul, Korea
(final version received 24 July 2012)
A recent revision of the local accounting standard for property, plant, and equipment
Downloaded by [Umeå University Library] at 13:24 03 October 2013
(PP&E) has conferred Korean companies a revaluation option for PP&E without the
early adoption of International Financial Reporting Standards, which will be manda-
tory in 2011. The stock market generally reacts favorably to the announcement of an
asset revaluation. We examine the motives and characteristics of companies that
revalue PP&E and their choice of assets for revaluation. First, we find that Korean
companies are more likely to revalue PP&E to improve their financial position or
reduce debt contracting costs rather than lessen political costs or signal better future
prospects. Second, we report a pecking order in the choice of assets to revalue.
Companies prefer land for revaluation to other depreciable assets. When companies
select depreciable assets for revaluation, most of them revalue land at the same time
suggesting that companies are likely to select depreciable assets for revaluation only
after they select land. Third, we posit and find that revaluing companies are more
likely to elect depreciable classes of PP&E when they are highly levered, experience
equity depletion, and report losses. We also find that revaluing companies are more
likely to recognize revaluation decrements in addition to increments when they are
large and are revaluing depreciable classes of PP&E.
Keywords: asset revaluations; revaluation surplus; PP&E; revaluation of depreciable
assets; revaluation of land; revaluation increments; revaluation decrements
JEL Code: M41
1. Introduction
The Korean Accounting Standards Board (KASB) has recently revised local accounting
standards to confer a revaluation option for property, plant, and equipment (PP&E).
Korean companies have been deprived of the revaluation option for PP&E since the
revocation of the Asset Revaluation Act in 2000. A recent hike in property prices and
the absence of the revaluation option for an extended period have increased the need
for asset revaluation. It has been alleged that the financial condition of Korean compa-
nies would appear less sound than that of companies in those countries that have
already adopted International Financial Reporting Standards (IFRS) or allowed a revalu-
ation in their local accounting standards. The KASB contends that the introduction of
PP&E revaluation facilitates a smooth transition to IFRS and makes the financial state-
ments of Korean companies more comparable to those of companies in the European
Union and Australia, which have already adopted IFRS. The revision of the local
accounting standard for PP&E has enabled Korean companies to take advantage of the
*Corresponding author. Email: jinhanpae@korea.ac.kr
Ó 2013 City University of Hong Kong and National Taiwan University

Asia-Pacific Journal of Accounting & Economics 145
benefits of asset revaluation without the early adoption of IFRS.1 Since the reintroduc-
tion of PP&E revaluation on 30 December 2008, about 18% of publicly listed Korean
companies have revalued PP&E in the first quarter of 2009.2 We examine the motives
and characteristics of the companies that revalue PP&E and their choice of items to
revalue.
We identify 302 listed Korean companies that adopted the revaluation model for
accounting periods that ended between December 2008 and March 2009, and examine
the determinants of the propensity to and the choice of items to revalue PP&E. We pos-
tulate that companies revalue PP&E when the expected benefits exceed the direct and
indirect costs of an asset revaluation (e.g. appraisal fees, audit fees, and the opportunity
costs for management for selecting items to revalue, reviewing the fair values, etc.).
First, we examine factors influencing a firm’s decision to revalue PP&E. Prior stud-
ies suggest that in the absence of tax benefits, companies mostly revalue fixed assets to
improve their financial position, reduce contracting and political costs, and signal better
future operating prospects (Brown et al. 1992, Whittred and Chan 1992, Easton et al.
1993, Cotter and Zimmer 1995, Lin and Peasnell 2000, Barlev et al. 2007). We find
that Korean companies are more likely to revalue PP&E when they need to improve
their financial position, avoid breaching debt covenants, or enhance their borrowing
capacity, but we do not find empirical evidence that is consistent with motives for
reducing political costs or signaling better future prospects.
Second, we examine the choice of items to revalue. Revaluations are neither manda-
tory nor required for all PP&Es. However, if an item of PP&E is revalued, companies
are required to revalue the entire class of PP&E to which that item belongs. Once com-
panies adopt the revaluation model, they need to decide which classes of PP&E to reva-
lue. We propose a pecking order in the choice of items to revalue. Revaluing
companies will prefer land to other depreciable assets because the upward revaluation
of land has no detrimental effect on future reported earnings. In our sample, all compa-
nies except one chose to revalue land and companies that opted to revalue depreciable
assets revalued land as well. The ratio of those revaluing both land and other deprecia-
ble assets to those revaluing only land is about 1:5. We hypothesize that companies are
more likely to select depreciable assets for revaluation when they are in great need of
an improved financial position or financing. Consistent with this prediction, we find that
companies are more likely to revalue depreciable classes of PP&E (mostly along with
land) when they are at risk of breaching the minimum legal capital covenant or when
they report losses.
Finally, we examine whether companies that recognize both revaluation decrements
and increments differ from those that recognize only revaluation increments. In our
sample, none of the revaluing companies report a net revaluation decrement. To some
extent, this asymmetry is expected because revaluations are voluntary and companies
have discretion over the timing of revaluation and the items to revalue. Revaluation
decrements are costly to revaluing companies because they decrease the reported earn-
ings. We contend that a revaluation decrement is an unintended consequence rather than
an outcome that revaluing companies set out to attain. Revaluation decrements may
arise when either the items in the chosen classes of PP&E are heterogeneous in their
changes in value or a variety of items are included in the classes chosen for revaluation.
While land prices have been increasing in recent years, this inflationary trend has been
much weaker or more varied for items in other classes of PP&E. Consistent with that,
we find that companies are more likely to recognize revaluation decrements when they
are large and are revaluing depreciable classes of PP&E.
146 T.H. Choi et al.
The main contribution of our study is twofold. First, even though many studies
explore the determinants of the propensity to revalue PP&E, only a few examine the
details of asset revaluations. We extend the literature on asset revaluation by proposing
a pecking order in the choice of assets for revaluation and examining why some compa-
nies revalue land and other depreciable assets at the same time when a majority of com-
panies revalue only land and why some companies recognize revaluation decrements as
well as revaluation increments when a majority of companies recognize only revaluation
increments. Our study sheds light on these unexplored venues of asset revaluation.
Second, we provide Korean evidence on the determinants of the propensity to reva-
lue PP&E. Barlev et al. (2007) report a variation in the factors that influence asset
revaluations across countries.3 Even though the recently adopted revaluation model is
similar to that available under IFRS, Korean companies differ from their peers in other
countries with respect to the motives and characteristics of their asset revaluations. For
example, the market-to-book ratio and size do not influence Korean companies’ propen-
sity to revalue PP&E once other relevant factors are controlled for, indicating that polit-
ical costs or growth opportunities are not the main determinants of the propensity of
Korean companies to revalue PP&E. Korean companies seem to be concerned about
maintaining the minimum legal capital to avoid delisting or enhancing the borrowing
capacity; this corroborates the importance of country-specific factors in understanding
asset revaluations by international companies.
The rest of the paper proceeds as follows. The next section explains accounting
rules for asset revaluation, develops a choice model for asset revaluation, and proposes
empirical hypotheses regarding the choice of items to revalue. Section 3 explains our
sample selection procedures and compares the revaluing and nonrevaluing companies.
Section 4 reports the estimation results for the determinants of the propensity to revalue
PP&E and a choice model for items to revalue and the reaction of the stock market to
the announcement of an asset revaluation. A summary follows along with concluding
remarks.
2. Accounting for asset revaluation and related research

2.1. Accounting for asset revaluation
The revised Korean accounting standard for PP&E is similar to International Account-
ing Standard 16 Property, Plant and Equipment (IAS 16).4 PP&E are initially recog-
nized at historical cost. After initial recognition, companies choose either the cost
model or the revaluation model for each class of PP&E. Under the cost model, PP&E
are carried at cost after the deduction of any accumulated depreciation and any accumu-
lated impairment losses. Under the revaluation model, PP&E are regularly updated at
fair value and the subsequent depreciation and impairment losses are based on the
revalued amount as of the date of revaluation. Depreciation and impairment losses are
entailed regardless of the method chosen for the subsequent valuation.
Accounting treatments for revaluation decrements and increments are asymmetric:
revaluation decrements are reported in earnings, whereas revaluation increments are
reported directly to equity, bypassing the income statement. Hence, companies cannot
increase reported earnings by revaluing PP&E. On the contrary, revaluation could be
detrimental to current and future earnings. When companies experience both revaluation
increments and decrements, they will experience a decrease in current earnings due to
the revaluation decrements even if the revaluation increments exceed the revaluation
decrements. If the revaluation increments are associated with depreciable assets, they
will result in a decrease in future earnings. In contrast, there is no detrimental deprecia-

tion effect for land, which partly explains why many companies choose land for
revaluation.
With respect to the frequency of PP&E revaluation, companies that chose the
revaluation option for PP&E are not required to revalue PP&E every year unless
there is a significant change in fair values. This rule is similar to that in IAS 16. In
Appendix, as an illustration, we present the PP&E revaluation made by Namsun
Aluminum Co., Ltd.
The revaluation model under the revised accounting standard, to some extent, differs
from the one available to Korean companies under the Asset Revaluation Act, which
was first enacted in 1965 and annulled in 2000. Under the Asset Revaluation Act, com-
panies could selectively revalue fixed assets and were required to pay an asset revalua-
tion tax for the appreciation in value. In return, more depreciation expenses were tax
deductible in the subsequent periods. In general, the net effect (tax savings due to
increased depreciation and net of the asset revaluation tax) was rather favorable. In con-
trast, changes in value that arise from asset revaluation under the revised accounting
standard are not recognized under Korean tax laws; hence, companies neither bear the
cost of the revaluation tax nor get the benefit of future tax savings that arise from asset
revaluation.
2.2. Determinants of the propensity to revalue PP&E

Companies will revalue fixed assets when the anticipated benefits of the revaluation
exceed the related costs (Whittred and Chan 1992). The costs of revaluation encompass
not only actual out-of pocket costs, such as appraisal fees when an independent valuer
is employed and an increase in audit fees charged by the auditors, but also record keep-
ing costs and opportunity costs, such as the time and effort managers spend in selecting
items to revalue, reviewing the fair values of the chosen items, and discussing these
values with the auditors (Brown et al. 1992, Henderson and Goodwin 1992). In devel-
oping a model of the determinants of the propensity to revalue PP&E, we take into
account local institutional and legal requirements as well as factors that were found to
be relevant in other countries, such as Australia, Belgium, and the UK (Brown et al.
1992, Whittred and Chan 1992, Cotter and Zimmer 1995, Gaeremynck and Veugelers
1999, Lin and Peasnell 2000, Barlev et al. 2007).
2.2.1. Intensities of PP&E and property

The revised accounting standard allows a revaluation for only PP&E.5 The potential
benefits arising from revaluation would be greater for companies with more PP&E
(Brown et al. 1992). Since PP&E is often provided as collateral, the revaluation of
PP&E will enhance a company’s borrowing capacity (Cotter and Zimmer 1995). We
predict that companies with higher ratios of PP&E to total assets (higher PP&E inten-
sity) are more likely to revalue PP&E than those with lower PP&E intensity. Among
various PP&E classes, property, such as land and buildings, is likely to generate more
benefits than other classes of PP&E, such as machinery and vehicles. As property prices
have been increasing in recent years, the disparity between the book value and the fair
value will be greater for property than for other PP&E items (Brown et al. 1992). We
thus predict that revaluation is an increasing function of the proportion of PP&E that is
property (namely, the property intensity).
2.2.2. Financial leverage

Whittred and Chan (1992) and Brown et al. (1992) report that Australian companies are
more likely to revalue PP&E when their financial leverage is high.6 This is consistent
with a survey finding of Easton et al. (1993) that Australian companies revalue PP&E
mainly to lower financial leverage. Considering that an asset revaluation leads to an
improvement in financial leverage, revaluation will lower the contracting costs of com-
panies that are close to their debt constraints. It is also consistent with the position of
the KASB that the revaluation of PP&E helps to improve the financial leverage of com-
panies with higher proportions of PP&E. We employ the ratio of total liabilities to total
assets as a measure of the financial leverage.
2.2.3. Signaling future performance

There are two opposing views regarding the information or signal that is conveyed by
the choice of revaluation.7 According to the signaling theory, a signal is credible only
when it is costly to convey the signal (Spence 1973). In our context of an asset revalua-
tion, revaluing and nonrevaluing companies face different types of signaling costs. Lin
and Peasnell (2000) and Barlev et al. (2007) argue that by revaluing PP&E, companies
can signal more favorable future prospects. The revaluation of PP&E is a credible sig-
nal for future performance because it is costly to the revaluing companies. Asset revalu-
ations are detrimental to the current and future returns on total assets (ROA) and the
current reported earnings. The current and future ROAs decrease due to an increase in
the asset base (the denominator effect) and future ROAs decrease due to the increased
depreciation charges when depreciable assets are revalued upward (the numerator
effect). The current earnings also decrease when the revaluation gives rise to a decrease
in value for some items in the chosen class of PP&E. Hence, Lin and Peasnell (2000)
and Barlev et al. (2007) predict that companies that anticipate better future performance
can afford to revalue, but those that anticipate poor future performance cannot afford to
revalue PP&E.
On the other hand, Gaeremynck and Veugelers (1999) argue that revaluation is a
negative signal of poor performance or closeness to debt constraints. Unsuccessful com-
panies cannot afford to forego the opportunity to revalue PP&E when they need to
improve their financial position to enhance their borrowing capacity or to avoid the vio-
lation of debt covenants. Companies with strong financial position can distinguish them-
selves from companies with weak financial position by electing not to revalue PP&E.
Thus, Gaeremynck and Veugelers (1999) predict that companies with strong financial
position are less likely to revalue PP&E.
We use the return on total assets before revaluation to represent the current operat-
ing performance and assume that the current operating performance partly portends
future operating performance. Due to the conflicting views about the nature of informa-
tion that is conveyed by the choice of revaluation, we do not make any prediction
regarding the relationship between the propensity to revalue PP&E and the return on
total assets before revaluation.
2.2.4. Political costs

Accounting information is often used in the political process. Large reported earnings
may be used as evidence of a monopoly and reported earnings are often used to set
rates in regulated industries such as public utilities. Under the assumption that large
companies are politically more sensitive than small companies (Watts and Zimmerman
1986), the size hypothesis suggests that large companies are more likely to revalue in
order to project a conservative picture of profitability, which will help to lower the
attention of the press and government. Consistent with the size hypothesis, prior studies
report a positive relationship between the propensity to revalue PP&E and size (Brown
et al. 1992, Lin and Peasnell 2000, Barlev et al. 2007).
2.2.5. Equity depletion

One of the requirements for continual listing on the Korea Exchange (KRX) is to main-
tain a minimum legal capital. Specifically, to avoid delisting, the rate of legal capital
depletion, which is defined as ((legal capital total equity)/(legal capital)), should be
maintained at less than 50%. Since delisting is very costly, companies may elect to
revalue PP&E to meet this legal capital retention covenant. We use an indicator variable
to represent whether or not companies breach the legal capital retention covenant before
revaluation.
2.2.6. Need for financing

Cotter and Zimmer (1995) argue that companies revalue PP&E to enhance their borrow-
ing capacity when they anticipate the need to raise capital in the future. Article 470 of
the Korean Commercial Act prohibits companies from issuing bonds that exceed four
times the net asset value. Companies that are close to this constraint and in need of
future financing will have incentives to increase their bond issuing capacity. The need
for additional capital will be great when companies fail to generate positive operating
cash flows or report losses. Hence, we predict that companies are more likely to revalue
PP&E when they expect to report negative operating cash flows (NegCFOD) or losses
prior to revaluation.
2.2.7. Past revaluation

Lin and Peasnell (2000) and Barlev et al. (2007) find that companies that revalued
fixed assets in the past are more likely to revalue them again. We include an indicator
variable to represent whether or not a company revalued fixed assets prior to 2000
under the Asset Revaluation Act. We predict that companies with prior revaluation
experiences are more likely to revalue PP&E than those without prior revaluation
experiences.
2.2.8. Market-to-book ratio

Whittred and Chan (1992) find a positive relationship between the propensity to revalue
PP&E and the market-to-book ratio for Australian companies. They infer that compa-
nies with more investment opportunities reduce the problems of underinvestment by
revaluing assets. On the other hand, Lin and Peasnell (2000) argue that the relationship
between the propensity to revalue PP&E and the market-to-book ratio could be negative
because the market-to-book ratio can be viewed as a proxy for undervaluation and a
revaluation alleviates the undervaluation problem in the capital market. Lin and Peasnell
(2000) report a negative relationship for British companies. Barlev et al. (2007) also
report a negative relationship for companies under the British accounting system, but
find no such relation for companies in other countries. Hence, we offer no prediction
regarding the relationship between the propensity to revalue PP&E and the market-to-
book ratio.
We summarize our discussions on the determinants of the propensity to revalue
PP&E in the following hypothesis.
Hypothesis 1: The propensity to revalue PP&E increases with the intensities of PP&E and
property, financial leverage, firm size, equity depletion, need for financing, and prior reval-
uation experiences.
2.3. Choice of Items to revalue

Asset revaluation rules do not require companies to revalue all PP&E. Companies have
discretion over PP&E classes to revalue. Prior studies (Brown et al. 1992, Whittred and
Chan 1992, Lin and Peasnell 2000, Barlev et al. 2007) suggest that companies are more
likely to revalue PP&E when the proportion of total assets that is PP&E or the propor-
tion of PP&E that is property is high. We extend this argument and propose a pecking
order in the choice of PP&E classes for revaluation: companies prefer land for revalua-
tion to other PP&E classes that are subject to depreciation.
Revaluing companies will take into account the consequences of the asset revalua-
tion when they decide on which PP&E classes to revalue. One of the consequences of
the upward revaluation of depreciable assets is an increase of the depreciation base,
which lowers future reported earnings through increased depreciation charges. By
choosing land, revaluing companies can increase the carrying amount of PP&E and net
assets without worrying about the detrimental effect of the increased depreciation base
on their future reported earnings. Thus, in the absence of tax incentives, we contend
that revaluing companies prefer land to depreciable classes of PP&E; therefore, for
revaluation, they select land ahead of depreciable assets.
Hypothesis 2: There is a pecking order in the choice of PP&E classes to revalue: compa-
nies are more likely to select land for revaluation ahead of depreciable classes of PP&E.
The pecking order conjecture (hypothesis 2) implies that fewer companies will revalue
depreciable assets, but companies that revalue depreciable assets are likely to also reva-
lue land at the same time. We are interested in the characteristics of those companies
who select depreciable classes of PP&E (mostly in addition to land). Since it is costly
to revalue depreciable classes of PP&E, companies will revalue such classes when they
are willing to bear reduced future profitability or expect to derive more benefits from
revaluation in spite of the increased future depreciation charges. We posit that compa-
nies are more likely to revalue depreciable classes of PP&E when they need to improve
their financial position or enhance their borrowing capacity. Thus, we predict that com-
panies are more likely to revalue depreciable classes of PP&E (mostly in addition to
land) when they have high financial leverage, experience equity depletion, and report
losses or negative cash flows.
Hypothesis 3: Companies are more likely to select depreciable assets for revaluation when
they need to improve their financial position or enhance their borrowing capacity.
3. Data and descriptive statistics

3.1. Sample selection and descriptive statistics
We collect accounting data from TS2000, which is a financial information database
maintained by the Korea Listed Companies Association. Our sample includes 1689
companies that are listed on the KRX and have filed annual financial statements with
the Korean Financial Supervisory Service for accounting periods that ended between
December 2008 and March 2009. We identify 302 revaluing companies as those with a
positive balance in the revaluation surplus on the balance sheet or those with nonzero
revaluation losses on the income statement.8
Panel A of Table 1 displays the descriptive statistics for the whole sample of 1689
listed companies. The mean (median) market capitalization is 362 billion Korean Won
(KRW) (32 billion KRW). The mean (median) shareholder’s equity is 377 billion KRW
(48 billion KRW). For the average listed Korean company, the book value of equity is
greater than the market value of equity: the mean market-to-book (MTB) ratio is less
than one – specifically, 0.984. The average total assets amount to 972 billion KRW.
Panel B displays the classes of PP&E that the revaluing companies selected. Land
is the most popular class. All companies, except one, revalued land. The next most pop-
ular class is buildings. Forty companies (13.2%) revalued buildings in addition to land.
Thirteen and ten companies revalued machinery and structures, respectively. The other
category of revaluation includes companies that revalued vehicles, electric equipment,
and a golf course (four companies). Note that the total number of companies exceeds
the number of revaluing companies because many companies revalued multiple classes
of PP&E.
Panel C presents a 2 2 table to report whether companies revalued only land or
whether they revalued other depreciable assets as well. There are three distinctive types
of revaluing companies: (i) those that revalued only land, (ii) those that revalued land
and some other PP&E, and (iii) those that revalued some PP&E other than land. In our
empirical analysis for the choice of items to revalue, we partition our revaluation sam-
ple into two groups: those that revalued only land and those that revalued PP&E other
than land. There are 249 companies (82.5%) that revalued only land and 53 companies
(17.5%) that revalued some depreciable classes of PP&E (along with land except for
one company).9
The results in Panels B and C indicate that there is a pecking order when companies
choose PP&E classes to revalue (Hypothesis 2). Companies prefer land for revaluation
to other classes of PP&E. When companies revalue assets other than land, most of them
(except for one in our sample) revalue land at the same time, suggesting that companies
are likely to select depreciable classes of PP&E for revaluation only after they select
land.
Panel D classifies the revaluing companies into those that reported revaluation
decrements and those that reported only revaluation increments. In our sample of
revaluing companies, 78 companies (25.8%) reported a revaluation decrement as well
as a revaluation increment, whereas 224 companies (74.2%) reported only revaluation
increments.10 Not a single company reported a revaluation decrement without reporting
a revaluation increment and the revaluation increment was always greater than the
revaluation decrement, indicating that companies revalue PP&E to recognize a
revaluation increment rather than a revaluation decrement.
Finally, Panel E divides the revaluation sample into companies that employed an
independent valuer in determining a fair value and those that did not employ an
Table 1. Descriptions of the sample.
Variables Mean Standard deviation First quartile Median Third quartile

Panel A: full sample – 1689 observations (in millions of KRW)
MVE 362,111 2,186,357 14,772 32,315 87,650
Sales 655,766 3,178,023 27,201 76,212 209,516
CFO 14,172 634,935 3,049 1,671 10,370
Equity 376,948 2,189,421 23,117 47,767 130,585
Liabilities 594,828 4,489,867 15,095 38,547 110,648
Assets 971,879 5,623,398 42,357 86,773 249,974
Net income 16,732 260,690 7,624 1,123 7,771
MTB 0.984 1.678 0.404 0.654 1.093
PP&E class Land Buildings Machinery Structures Others

Panel B: classification by PP&E classes (no. of revaluing companies = 302)

No. of companies 301 40 13 10 4
(Percentage) (99.7%) (13.2%) (4.3%) (3.3%) (1.3%)
Revaluation No revaluation
of depreciable of depreciable
items (other than land) items (other than land) Total
Panel C: revaluation of land vs. depreciable items (no. of revaluing companies = 302)
Revaluation of land 52 249 301 (99.7%)
No revaluation of land 1 N/A 1 (0.3%)
Total 53 (17.5%) 249 (82.5%) 302 (100.0%)
Revaluation No revaluation
decrements decrements Total
Panel D: revaluation increments vs. decrements (no. of revaluing companies = 302)
Revaluation increments 78 224 302 (100.0 %)
No revaluation increments 0 N/A 0 (0.0 %)
Total 78 (25.8 %) 224 (74.2 %) 302 (100.0 %)
Independent No independent
valuer valuer Total
Panel E: use of independent valuers (no. of revaluing companies = 302)
Revaluation of land only 232 17 249 (82.5%)
Revaluation of items other than land 53 0 53 (17.5%)
Total 285 (94.4%) 17 (5.6%) 302 (100.0%)
Notes: Our sample includes 1,689 companies that are listed on the KRX and filed their annual financial
statements for accounting periods that ended between December 2008 and March 2009. The three hun-
dred and two companies revalued some classes of PP&E. We collect the stock price at the end of the
accounting period and basic accounting data from TS2000, which is a financial information database
maintained by the Korea Listed Companies Association. Other detailed data on revaluations are hand-
collected from the annual reports, which are available at KIND (http://kind.krx.co.kr). MVE is the mar-
ket value of common equity. CFO denotes cash flows from operations. MTB is the market-to-book
ratio.
independent valuer. In our sample, the majority of revaluing companies (94.4%)

employed an independent valuer. The percentage of companies using an independent
valuer is much higher than the figure of 46% reported by Cotter and Richardson (2002)
for Australian revaluing companies.11 All companies that revalued PP&E other than
land employed independent valuers. Only 17 companies (5.6%) did not employ an
independent valuer. These companies have chosen to revalue only land and used the
officially assessed land prices (OALP), which the government announces each year and
uses to assess property-related taxes. One company employed an independent valuer
and reported the use of OALP at the same time, as there was not much difference
between the OALP and the value appraised by the independent valuer.
3.2. Impact of revaluation

An asset revaluation can give rise to both revaluation increments and decrements. The
revaluation increment is reported on an after-tax basis as an increase in the revaluation
surplus on the balance sheet, whereas the revaluation decrement is reported on a
before-tax basis on the income statement.12 Hence, we adjust the tax effect for the
revaluation decrement in order to assess the net effect of asset revaluation on changes
in equity.13
Panel A of Table 2 reports the effects of asset revaluations on the balance sheet and
net income. As a result of revaluation, assets increase, on average, by 68.3 billion
KRW or 16% of the amount before revaluation. For all the publicly listed companies,
the total increase in assets amounts to 20.6 trillion KRW. The average changes in
liabilities and equity are 14.8 billion KRW and 53.5 billion KRW, respectively. They
represent 5.8 and 82.9% of the corresponding amounts before revaluation. The average
decrease in the net income is 398 million KRW or 1.7% of the amount before
revaluation.14
In our sample, there are 22 companies with negative shareholders’ equity on a
before asset revaluation basis. With the help of an asset revaluation, 12 of them avoided
reporting negative shareholders’ equity; however, the remaining 10 companies could
not avoid reporting negative shareholders’ equity even after asset revaluation. Negative
shareholders’ equity has grave consequences because it is one of the causes of delisting.
Of the 10 companies that reported negative shareholders’ equity, five companies were
eventually delisted.
Panel B of Table 2 displays descriptive statistics of the revaluing companies before
and after revaluation and tests whether the revaluing companies differ from nonrevalu-
ing companies with respect to market capitalization, total sales, operating cash flows,
financial position, and net income. Asset revaluations do not influence the market capi-
talization, sales, and cash flows from operations of the revaluing companies because the
decision to revalue PP&E is made after the fiscal period end-date.15 There are no statis-
tically significant differences in the means and medians of market capitalization between
revaluing and nonrevaluing companies. Revaluing companies are, on average, less prof-
itable than nonrevaluing companies and more than half the revaluing companies do not
generate positive operating cash flows. It is interesting that before asset revaluation,
revaluing companies had smaller mean and median book values of equity than nonre-
valuing companies but after asset revaluation, there was no significant difference in the
median book values of equity (47,594 million KRW vs. 49,491 million KRW). As
explained above, PP&E revaluations significantly increase the equity of revaluing
companies.
Next, we explore how the stock market reacts to the announcement of an asset
revaluation for a sample of 90 firms that disclose their plans to revalue PP&E and the
results of asset revaluation on a voluntary basis before they announce earnings or issue
financial statements.16 We expect a favorable reaction by the stock market to the
announcement of an asset revaluation, because companies will revalue PP&E when the
Table 2. The impact of asset revaluation and comparison of revaluing vs. nonrevaluing
companies.
Variables Mean Standard deviation First quartile Median Third quartile

Panel A: effect of revaluation in millions of KRW (% change)
Change in assets 68,292 174,205 6,530 15,415 38,850
(16.0%) (6.1%) (11.7%) (21.3%)
Change in liabilities 14,784 38,415 986 3237 8379
(5.8%) (1.2%) (3.1%) (6.5%)
Change in equity 53,508 135,812 5477 12,847 31,063
(82.9%) (16.0%) (31.2%) (64.2%)
Change in net income 398 2091 3 0 0
( 1.7%) ( 0.0%) (0.0%) (0.0%)
Note: % changes are relative to the amount before revaluation.

Tests of differences
between
non-revaluing and
Revaluing revaluing
companies (N = 302) companies (p-values)
Non-revaluing After Before After Before
companies (N = 1387) revaluation revaluation revaluation revaluation
Mean Mean Mean Mean Mean
Variables (Median) (Median) (Median) (Median) (Median)
Panel B: tests of differences between nonrevaluers and revaluers (in millions of KRW)
MVE 398,511 194,934 No change 0.006 No change
(32,347) (30,753) (0.300)
Sales 647,112 695,513 No change 0.763 No change
(70,853) (112,262) (<0.001)
CFO 24,640 33,906 No change 0.203 No change
(2,175) ( 443) (<0.001)
Equity 403,750 253,855 200,347 0.048 0.005
(47,594) (49,491) (35,847) (0.562) (0.001)
Liabilities 547,831 810,669 795,885 0.451 0.476
(32,544) (70,787) (67,208) (<0.001) (<0.001)
Assets 951,707 1,064,524 996,232 0.771 0.908
(81,289) (130,247) (109,460) (<0.001) (<0.001)
Net income 22,564 10,054 9655 0.002 0.002
(1632) ( 3690) ( 3628) (<0.001) (<0.001)
statements for accounting periods that ended between December 2008 and March 2009. 302 companies
revalued some classes of PP&E. We collect the stock price at the end of the accounting period and
basic accounting data from TS2000, which is a financial information database maintained by the Korea
Listed Companies Association. Other detailed data on revaluations are hand-collected from the annual
reports, which are available at KIND (http://kind.krx.co.kr). MVE is the market value of common
equity. CFO denotes cash flows from operations.
We estimate changes in assets, liabilities, equity, and the net income as follows:
• Increase in assets = (after-tax) revaluation increment + deferred tax liability for the revaluation
increment – (before-tax) revaluation decrement.
• Increase in liabilities = deferred tax liability for the revaluation increment – [(before-tax) revalua-
tion decrement deferred tax rate].
• Increase in equity = (after-tax) revaluation increment – [(before-tax) revaluation decrement (1 –
deferred tax rate)].
• Decrease in net income = (before-tax) revaluation decrement (1 – deferred tax rate).
In the above, the deferred tax rate = deferred tax liability for the revaluation increment/(after-tax
revaluation increment + deferred tax liability for the revaluation increment).
Table 3. Market reaction to the announcement of an asset revaluation.
Event window ( 1 to +1) ( 2 to +2) ( 3 to +3)

Panel A: CAR of revaluing companies (N = 90)
Cumulative abnormal return 0.0483 0.0501 0.0587
(t-statistic) (5.05)⁄⁄ (4.15)⁄⁄ (4.43)⁄⁄
⁄⁄
Significant at the 1% level.
Event window ( 1 to +1) ( 2 to +2) ( 3 to +3)

Panel B: comparison of the CARs between revaluing and non-revaluing companies (N = 85)
CAR: revaluing firms (A) 0.0489 0.0498 0.0597
CAR: nonrevaluing firms (B) 0.0315 0.0447 0.0517
Difference (A B) 0.0174 0.0051 0.0080
(t-statistics) (1.25) (0.29) (0.43)

Notes: The sample includes 90 firms that voluntarily disclosed their plans to revalue PP&E and those
that voluntarily disclosed the results of revaluation. The cumulative abnormal stock returns are calcu-
lated around the announcement of an asset revaluation. The abnormal stock returns are calculated by
subtracting value-weighted market returns from raw stock returns. In Panel B, for each revaluing firm,
a matching nonrevaluing firm in the same industry.
benefits of revaluation exceed the costs. Revaluation will help companies to increase
their borrowing capacity by improving the financial leverage ratio and to avoid delisting
by ameliorating the degree of equity depletion. On the other hand, it is also possible
that the stock market interpret an announcement of an asset revaluation as a signal for a
weak financial position, because financially strained companies are more likely to reva-
lue PP&E than financially sound companies.
Panel A of Table 3 reports the cumulative abnormal returns (CAR) around the
announcement of an asset revaluation: 4.83% for the three-day window ( 1 to +1),
5.01% for the five-day window ( 2 to +1), and 5.87% for the seven-day window ( 3
to +3). Significantly positive three-day CAR imply that the stock market takes an asset
revaluation as a favorable signal to revaluing firms. Panel B of Table 8 reports the
differences in the CAR between the revaluing and nonrevaluing firms.17 Revaluing
firms exhibit higher CARs than non-revaluing firms. For example, the CAR for the
three-day window ( 1 to +1) is 4.89% for revaluing companies and 3.15% for nonre-
valuing companies. But, the mean difference of the CARs (1.74%) is not statistically
significant. We have similar patterns for other event windows. To summarize, the posi-
tive market reaction to the announcement of an asset revaluation and a higher average
CAR for revaluing companies suggest that firms generally enjoy net benefits from asset
revaluations.
4. Empirical results
4.1. Univariate tests of revaluing vs. non-revaluing companies
Table 4 presents the univariate test results for the differences in means and medians
between revaluing and nonrevaluing companies for the 10 variables that are hypothe-
sized to affect the propensity to revalue PP&E. For revaluing companies, we remove
the effects of a revaluation to facilitate a comparison with the nonrevaluing companies.
That is, all variables are presented on a before-revaluation basis, where applicable. The
expected signs are presented in parentheses. We predict and find that the PP&E and
Table 4. Univariate tests of variables affecting the propensity to revalue PP&E.
Revaluing Non-revaluing
companies companies
(N = 302) (N = 1,387) Tests of differences in
Variable (expected sign for the Mean Median
test of revalues nonrevaluers) Mean Median Mean Median (t-statistic) (Z-statistic)
PP&E intensity (+) 0.303 0.276 0.232 0.210 (6.30)⁄⁄⁄ (6.48)⁄⁄⁄
Property intensity (+) 0.749 0.769 0.628 0.712 (6.41)⁄⁄⁄ (4.07)⁄⁄⁄
Leverage (+) 0.725 0.719 0.450 0.406 (15.73)⁄⁄⁄ (17.42)⁄⁄⁄
ROA (?) 0.159 0.029 0.169 0.016 (0.32) ( 7.72)⁄⁄⁄
Size (+) 18.908 18.511 18.551 18.214 (3.66)⁄⁄⁄ (4.23)⁄⁄⁄
EqDepletD (+) 0.109 0.000 0.048 0.000 (3.23)⁄⁄⁄ (4.07)⁄⁄⁄
NegCFOD (+) 0.536 1.000 0.373 0.000 (5.17)⁄⁄⁄ (5.23)⁄⁄⁄
(6.08)⁄⁄⁄ (6.09)⁄⁄⁄
LossD (+) 0.573 1.000 0.382 0.000

PastRevalD (+) 0.354 0.000 0.249 0.000 (3.53)⁄⁄⁄ (3.75)⁄⁄⁄
MTB (?) 0.818 0.719 0.985 0.669 (0.66) (0.95)
⁄⁄⁄
Significant at the one percent level in a two-tailed test.
dred and two companies revalued some classes of PP&E. Variables, where applicable, are measured on
a before-revaluation basis. PP&E intensity is the ratio of PP&E to total assets. Property intensity is the
proportion of PP&E that is property. Leverage is the ratio of total liabilities to total assets. ROA is the
return on total assets. Size is the logarithm of total assets. EqDepletD is an indicator variable equal to
one if the rate of legal-capital depletion, defined by ((legal capital total equity)/(legal capital)), is at
least 50%. NegCFOD is an indicator variable for companies with NegCFOD. LossD is an indicator var-
iable for companies that reported losses. PastRevalD is an indicator variable for companies with past
revaluation experiences. MTB is the market-to-book ratio.
property intensities are on average greater for revaluing companies than for nonrevalu-
ing companies. The test results for the means and medians with regard to leverage, size,
and dummies for equity depletion, negative cash flows, and losses are also consistent
with our predictions. However, we do not offer predictions for the ROA and the MTB
ratio due to conflicting views or findings in the extant literature. In fact, we find no sig-
nificant difference in the mean ROA and MTB between revaluing and nonrevaluing
companies, although the median ROA is significantly lower for revaluing companies
than for nonrevaluing companies. Overall, the univariate test results are consistent with
our predictions.
Table 5 presents Pearson correlation coefficients for the variables that are used in
the probit regressions for the propensity to revalue PP&E. Overall, the correlations
between the indicator variable for PP&E revaluation (RevalD) and the hypothesized
determinants of revaluation are of the predicted signs and significant at the one-percent
level. Prior studies reported mixed evidence on the effect of ROA and MTB on the pro-
pensity to revalue PP&E. In our sample of Korean companies, ROA and MTB are not
significantly associated with the adoption of the revaluation model.
We note that several variables are highly correlated with each other. For example,
the incidence of equity depletion (EqDepletD) is highly positively correlated with lever-
age (0.547) and highly negatively correlated with ROA ( 0.577). The correlation
between leverage and ROA is also very high and negative ( 0.598). In addition, not
surprisingly, the incidence of reporting NegCFOD and the incidence of reporting losses
(LossD) are highly positively correlated (0.353). In the sensitivity tests, we check the
robustness of our results when excluding these variables.
Table 5. Correlations between determinants of the propensity to revalue (p-values).
PP&E intensity Property intensity Leverage ROA Size EqDepletD NegCFOD LossD PastRevalD MTB
RevalD 0.149 0.125 0.282 0.005 0.087 0.099 0.127 0.148 0.091 0.028
(<0.01) (<0.01) (<0.01) (0.82) (<0.01) (<0.01) (<0.01) (<0.01) (<0.01) (0.25)
PP&E intensity 0.176 0.082 0.076 0.099 0.009 0.164 0.056 0.203 0.056
(<0.01) (<0.01) (<0.01) (<0.01) (0.71) (<0.01) (0.02) (<0.01) (0.02)
Property intensity 0.028 0.143 0.077 0.058 0.019 0.134 0.172 0.083
(0.26) (<0.01) (<0.01) (0.02) (0.44) (<0.01) (<0.01) (<0.01)
Leverage 0.598 0.028 0.547 0.175 0.271 0.050 0.026
(<0.01) (0.24) (<0.01) (<0.01) (<0.01) (0.04) (0.29)
ROA 0.313 0.577 0.240 0.375 0.131 0.020
(<0.01) (<0.01) (<0.01) (<0.01) (<0.01) (0.40)
Size 0.205 0.146 0.322 0.260 0.014
(<0.01) (<0.01) (<0.01) (<0.01) (0.57)
EqDepletD 0.178 0.297 0.112 0.045
(<0.01) (<0.01) (<0.01) (0.07)
NegCFOD 0.353 0.065 0.006
(<0.01) (0.01) (0.80)
LossD 0.158 0.021
(<0.01) (0.38)
PastRevalD 0.063
(0.01)
Notes: Our sample includes 1,689 companies that are listed on the KRX and filed their annual financial statements for accounting periods that ended between December
2008 and March 2009. The three hundred and two companies revalued some classes of PP&E. Variables, where applicable, are measured on a before-revaluation basis.
PP&E intensity is the ratio of PP&E to total assets. Property Intensity is the proportion of PP&E that is property. Leverage is the ratio of total liabilities to total assets.
ROA is the return on total assets. Size is the logarithm of total assets. EqDepletD is an indicator variable equal to one if the rate of legal-capital depletion, defined by
Asia-Pacific Journal of Accounting & Economics
((legal capital total equity)/(legal capital)), is at least 50%. NegCFOD is an indicator variable for companies with NegCFOD. LossD is an indicator variable for com-
panies that reported losses. PastRevalD is an indicator variable for companies with past revaluation experiences. MTB is the market-to-book ratio.
157
Table 6. Probit regression – determinants of the propensity to revalue PP&E.

158
Model (1) Model (2) Model (3) Model (4)

Expected Coefficient Coefficient Coefficient Coefficient
Variable sign (p-value) dF/dx (p-value) dF/dx (p-value) dF/dx (p-value) dF/dx
Intercept ? 3.026 2.276 4.813 4.099
(<0.01) (<0.01) (<0.01) (<0.01)
PP&E intensity + 0.902 0.169 0.873 0.151 1.062 0.236 1.140 0.257
(<0.01) (<0.01) (<0.01) (<0.01)
Property intensity + 0.655 0.122 0.645 0.112 0.678 0.151 0.639 0.144
(<0.01) (<0.01) (<0.01) (<0.01)
Leverage + 2.240 0.419 2.151 0.372 0.808 0.180 0.900 0.202
(<0.01) (<0.01) (<0.01) (<0.01)
ROA ? 1.558 0.291 5.991 1.037
(<0.01) (<0.01)
Size + 0.008 0.002 0.026 0.004 0.122 0.027 0.090 0.020
(0.79) (0.41) (<0.01) (<0.01)
EqDepletD + 0.317 0.070 0.410 0.088
(0.22) (0.12)
NegCFOD + 0.337 0.066 0.263 0.047 0.309 0.071 0.403 0.094
T.H. Choi et al.
(<0.01) (<0.01) (<0.01) (<0.01)

LossD + 0.547 0.109 0.238 0.042 0.388 0.089
(<0.01) (0.04) (<0.01)
ROA LossD ? 7.578 1.311
(<0.01)
PastRevalD + 0.264 0.053 0.262 0.049 0.229 0.054 0.198 0.047

(<0.01) (0.01) (0.01) (0.02)
MTB ? 0.010 0.002 0.008 0.001 0.002 0.001 0.004 0.001
(0.56) (0.64) (0.88) (0.80)
Wald chi-square 242.57 251.18 162.52 137.4
(p-values) (<0.01) (<0.01) (<0.01) (<0.01)
Notes: Our sample includes 1,689 companies that are listed on the KRX and filed their annual financial statements for accounting periods that ended between December
2008 and March 2009. The three hundred and two companies revalued some classes of PP&E. Independent variables, where applicable, are measured on a before-reval-
uation basis. PP&E intensity is the ratio of PP&E to total assets. Property Intensity is the proportion of PP&E that is property. Leverage is the ratio of total liabilities to
total assets. ROA is the return on total assets. Size is the logarithm of total assets. EqDepletD is an indicator variable equal to one if the rate of legal-capital depletion,
defined by ((legal capital total equity)/(legal capital)), is at least 50%. NegCFOD is an indicator variable for companies with NegCFOD. LossD is an indicator variable
for companies that reported losses. PastRevalD is an indicator variable for companies with past revaluation experiences. MTB is the market-to-book ratio. The dF/dx
column reports changes in the probability of revaluing PP&E when each variable changes: an infinitesimally small amount for continuous variables and from 0 to 1 for
indicator variables.
The p-values in parentheses are from a two-tailed test of z-statistics based on robust standard errors.
dF/dx represents the marginal effect on the change in the probability of revaluing PP&E for an infinitesimal change in continuous independent variables or a change
from 0 to 1 in indicator variables.
Asia-Pacific Journal of Accounting & Economics
159
4.2. Determinants of the propensity to revalue PP&E (hypothesis 1)

Table 6 presents the estimation results of probit models for the propensity to revalue
PP&E. We present the estimation results of four probit models. Model (1) includes 10
variables that are hypothesized to influence a company’s choice of revaluation model
for some classes of PP&E. Model (2) further includes an interaction term for ROA and
LossD. We include the interaction term in Model (2) to allow for different coefficients
on ROA between profitable and loss firms. In contrast, Model (1) implicitly assumes an
equal coefficient on ROA. The incentives associated with asset revaluation may differ
between companies that report profits (profitable companies) and companies that report
losses (loss-making companies): more specifically, the arguments for political costs or
the signaling of future performance are better posed for profitable companies than for
loss-making companies. Thus, it would be inappropriate to assume that the impact of
the firm profitability on the propensity to revalue PP&E is the same between profitable
firms and loss firms. In Models (3) and (4), we exclude some variables that are found
to be highly correlated (see Table 5), namely, ROA, EqDepletD, and LossD.
We present the estimated coefficients for the probit models and the marginal effects
of each explanatory variable. The dF/dx column reports changes in the probability of
revaluing PP&E when each variable changes. For continuous variables such as PP&E
Intensity, Property Intensity Leverage, ROA, size, and MTB, we assume an infinitesi-
mally small change, whereas for indicator variables such as EqDepletD, NegCFOD,
LossD, and PastRevalD, we assume a change in the status of the indicator variable, that
is, from 0 to 1. The p-values in parentheses are from a two-tailed test of z-statistics
based on the robust standard errors.
Companies are more likely to revalue PP&E when they have more PP&E (in terms
of the PP&E Intensity) and a greater proportion of PP&E that is property (Property
Intensity). In Model (1), the estimated coefficients on PP&E Intensity and Property
Intensity are significant at the one-percent level and their marginal effects on the proba-
bility of revaluing PP&E are 0.169 and 0.122, respectively. We can interpret the results
as follows. Table 4 reports that for revaluing companies, the mean values of PP&E
Intensity and Property Intensity before revaluation are 0.303 and 0.749, respectively,
and the corresponding ratios for nonrevaluing companies are 0.232 and 0.628, respec-
tively. When PP&E Intensity increases from 0.2 to 0.3 (the approximate mean differ-
ence between revaluing and nonrevaluing companies), the probability of revaluing
PP&E increases by 1.69%. Even after controlling for the proportion of total assets that
is accounted for by PP&E, the propensity to revalue PP&E increases as the proportion
of PP&E that is property increases. If Property Intensity increases from 0.6 to 0.7, com-
panies are 1.22% more likely to revalue PP&E. Our results suggest that, all else being
equal, companies with more eligible assets for revaluation and more property are more
likely to revalue PP&E because the potential benefits from revaluation would be
greater.
Companies are more likely to revalue PP&E when their financial leverage before
revaluation is greater. The estimated coefficient on financial leverage before revaluation
is significant at the one-percent level across all models. In Model (1), the marginal
effect of financial leverage is 0.419. Note that revaluing companies are more levered
than nonrevaluing companies: the mean leverage before revaluation is 0.725 for revalu-
ing companies and 0.450 for nonrevaluing companies (Table 4). The higher mean finan-
cial leverage contributes to an estimated increase of 11.5% (0.419 (0.725 0.450)) in
the probability that a typical revaluing company will revalue PP&E.
Table 7. Choice of items to revalue – land vs. depreciable assets.
Companies
revaluing Companies
depreciable assets revaluing only land
(N = 53) (N = 249) Tests of differences in
Mean Median
Variable (expected sign) Mean Median Mean Median (t-statistic) (Z-statistic)
Panel A: univariate tests of differences in means and medians
Land intensity (–) 0.330 0.332 0.341 0.315 ( 0.41) ( 0.37)
Leverage (+) 0.822 0.835 0.704 0.704 (2.98)⁄⁄⁄ (3.75)⁄⁄⁄
ROA (?) 0.289 0.147 0.131 0.009 ( 2.52)⁄⁄ ( 4.27)⁄⁄⁄
Size (+) 18.652 18.409 18.963 18.515 ( 1.36) ( 1.52)
EqDepletD (+) 0.321 0.000 0.064 0.000 (3.85)⁄⁄⁄ (5.42)⁄⁄⁄
NegCFOD (+) 0.585 1.000 0.526 1.000 (0.78) (0.78)
LossD (+) 0.792 1.000 0.526 1.000 (4.13)⁄⁄⁄ (3.55)⁄⁄⁄
PastRevalD (+) 0.264 0.000 0.373 0.000 ( 1.60) ( 1.51)

MTB (?) 0.201 0.664 1.035 0.736 ( 1.35) ( 0.86)
⁄⁄⁄ (⁄⁄)
Significant at the one (five) percent level in a two-tailed test.
Controlling for the selection bias Single probit regression

Variable Expected sign Coefficient (p-value) Coefficient (p-value)
Panel B: probit regression for the election of depreciable classes of PP&E for revaluation
Intercept ? 2.876 2.256
(<0.01) (0.09)
Land intensity – 0.103 0.155
(0.79) (0.75)
Leverage + 1.302 0.128
(<0.01) (0.80)
ROA ? 2.104 2.740
(0.46) (0.54)
Size + 0.001 0.044
(0.98) (0.54)
EqDepletD + 1.177 1.327
(<0.01) (<0.01)
NegCFOD + 0.106 0.039
(0.48) (0.84)
LossD + 0.594 0.630
(<0.01) (0.01)
ROALossD ? 3.414 2.212
(0.23) (0.62)
PastRevalD + 0.018 0.120
(0.91) (0.56)
MTB ? 0.018 0.015
(0.23) (0.40)
Wald chi-square (p-values) 85.6 (<0.01) 32.91 (<0.01)

Test of independence with the 20.08 (<0.01) N/A
selection model: chi-square
(p-values)
Notes: The p-values in parentheses are from a two-tailed test of z-statistics based on the robust standard
errors.
Our sample includes 1,689 companies that are listed on the KRX and filed their annual financial state-
ments for accounting periods that ended between December 2008 and March 2009. The three hundred
and two companies revalued some classes of PP&E. Independent variables, where applicable, are mea-
sured on a before-revaluation basis. Land intensity is the proportion of PP&E that is land. Leverage is
the ratio of total liabilities to total assets. ROA is the return on total assets. Size is the logarithm of total
assets. EqDepletD is an indicator variable equal to one if the rate of legal-capital depletion, defined by
((legal capital total equity)/(legal capital)), is at least 50%. NegCFOD is an indicator variable for com-
panies with NegCFOD. LossD is an indicator variable for companies that reported losses. PastRevalD is
an indicator variable for companies with past revaluation experiences. MTB is the market-to-book ratio.
We do not offer a prediction on the ROA because there are opposing theoretical
arguments with regard to the signaling content of revaluation and because extant studies
report mixed empirical findings. Our empirical results for ROA are sensitive to the
adopted model. In Model (1), the estimated coefficient on ROA is significantly positive,
whereas it is significantly negative in Model (2). Model (1) assesses the impact on the
propensity to revalue PP&E for all companies in the sample, whereas by including an
interaction term, namely, ROA LossD, Model (2) allows us to assess the effect of
profitability separately between profitable and loss-making firms.
We argue that the costly credible signaling arguments for revaluation are better
posed for companies that report profits (profitable companies) than for companies that
report losses (loss-making companies), because signaling about better future perfor-
mance or better financial health is more credible for profitable companies than for loss-
making companies. In that case, Model (2) will be better specified than Model (1).
Revaluations will be more beneficial to loss-making companies than to profitable com-

panies if the former have a higher leverage than the latter. The average leverage is
0.617 for loss-making companies and 0.414 for profitable companies (this is not tabu-
lated). The negative estimated coefficient on ROA in Model (2) suggests that companies
are less likely to revalue PP&E when they have a higher ROA. Nonrevaluing compa-
nies may signal better future prospects or better financial health by electing not to reva-
lue PP&E or foregoing the potential benefits of revaluation, such as improved financial
leverage. Our finding of a negative association between the propensity to revalue PP&E
and ROA for profitable companies is consistent with the prediction of Gaeremynck and
Veugelers (1999) rather than that of Lin and Peasnell (2000) and Barlev et al. (2007).
Prior studies generally report a positive coefficient on size. The positive coefficient
is interpreted as follows: companies that are politically more sensitive (that is, large
firms) are more likely to revalue PP&E in order to reduce the potential political costs.
In Models (1) and (2), our estimation results do not support the size (or political cost)
hypothesis. However, in Models (3) and (4), where some highly correlated variables are
excluded, the coefficient estimate for size becomes significantly positive. It is possible
that the high correlation of size with other explanatory variables in Models (1) and (2)
renders size insignificant in those models. However, size is a noisy proxy for political
costs and the results for ROA for profitable companies in Model (2) do not support the
political cost argument that profitable companies are more likely to revalue PP&E to
reduce current and future ROAs. To the extent that the high correlations between vari-
ables in Models (1) and (2) do not significantly distort the estimation results, the differ-
ences in size do not seem to explain Korean companies’ propensity to revalue PP&E.
EqDepletD is an indicator variable for companies that experienced a depletion of
the legal capital by more than 50% prior to revaluation. Since delisting is very costly,
we predict that companies are more likely to revalue PP&E to meet the legal capital-
retention covenant. The estimated coefficients on EqDepletD are of the predicted sign
but insignificant at the conventional level. We do not find a significant association
between the propensity to revalue PP&E and the incidence of equity depletion. Since
companies experiencing equity depletion are likely to have high leverage, we interpret
that the effect of high leverage subsumes the effect of equity depletion.
The need for financing will be greater for companies that report NegCFOD or losses
than for those that report positive operating cash flows or profits. Companies may
enhance their borrowing capacity by revaluing PP&E. Consistent with this argument,
NegCFOD and LossD are significantly positive at the one-percent level. In Model (2),
the companies that report NegCFOD are 4.7% more likely to revalue PP&E than
companies that report positive operating cash flows; further, companies that report
losses are 4.2% more likely to revalue PP&E than those that report profits. In our sam-
ple of listed Korean companies, 42% of companies reported losses and 40% of compa-
nies reported NegCFOD. Twenty-five percent of companies reported both NegCFOD
and losses. Only 43% of companies reported positive operating cash flows, but reported
profits. Compared to companies that report both profits and positive operating cash
flows, companies that report both losses and NegCFOD are 8.9% more likely to revalue
PP&E.
Consistent with prior research, the companies that revalued PP&E in the past are
more likely to revalue PP&E. In Model (2), companies with prior revaluation experi-
ences are about 5% more likely to revalue PP&E. We do not find a significant associa-
tion between the propensity to revalue PP&E and the MTB ratio, indicating no
significant association between the propensity to revalue PP&E and future investment
opportunities.
To summarize, our empirical results for the propensity of Korean companies to reva-
lue PP&E are generally consistent with our predictions and those of prior studies. The
propensity to revalue PP&E is an increasing function of the PP&E intensity, property
intensity, leverage, NegCFOD, losses, and past revaluations. The inferences about the
association between the ROA and the propensity to revalue PP&E are inconclusive;
however, we report that the association is negative at least for companies that report
profits. Equity depletion, size, and the MTB ratio do not exhibit a significant association
with the propensity to revalue PP&E.
4.3. Choice of Items to revalue

In this section, we examine whether companies that revalue only land differ from those
that revalue depreciable classes of PP&E (mostly in addition to land). As noted earlier,
Panel C of Table 1 shows that in our sample, 249 companies (82.5%) elected only land
for revaluation, whereas 53 companies (17.5%) elected depreciable classes of PP&E for
revaluation; further, with only one exception, companies in the latter group revalued
land at the same time.
The research question of the choice of assets to revalue focuses on the companies
that chose to revalue some of their PP&E. We are interested in which classes of PP&E
are chosen by revaluing companies and which classes of PP&E are left out. Note that
we cannot observe such choices for companies that did not choose to revalue. More-
over, companies that revalue PP&E differ from those who did not in their reporting
objectives and incentives as shown in the preceding section. That is, the revaluation
sample is not a random sample of the whole population. Thus, we need to control for
the sample selection bias of our revaluation sample. Failure to control for the sample
selection bias will lead to erroneous inferences. We employ the standard Heckman’s
selection bias adjustment approach (Greene 2003). We explicitly control for the selec-
tion of the revaluation model (as opposed to the cost model) by estimating the selection
model for the revaluation option for PP&E (Model 2 in Table 6). Our main probit
regression model for the election of depreciable classes of PP&E for revaluation is esti-
mated along with the probit regression model for the selection of the revaluation model
for some classes of PP&E.
The explanatory variables in the probit model for the election of depreciable classes
of PP&E for revaluation are similar to those in the selection model for the revaluation
option for PP&E (Model 2 in Table 6), except that the former probit model uses the
land intensity, as defined by the ratio of land to PP&E (instead of the PP&E and prop-
erty intensities). If companies with more land get enough benefits by revaluing only
land, they will have less incentive to revalue depreciable assets. We predict a negative
coefficient on land intensity.
Panel A of Table 7 presents the univariate test results for the differences between
companies that revalue depreciable assets (mostly in addition to land) and those that
revalue only land. There is no significant difference in land intensity between the two
groups of companies, indicating that the choice of depreciable classes of PP&E in addi-
tion to land is not mechanically associated with the ratio of land to PP&E. The inci-
dence of legal-capital depletion and losses before revaluation is significantly higher for
companies that elect depreciable assets for revaluation than for those that elect only
land for revaluation. However, there is no significant difference in size and past revalua-
tion experiences between the two groups.
Panel B presents the estimation results of the probit models for the election of
depreciable classes of PP&E for revaluation vs. the election of only land for revalua-
tion, both with and without control for the selection bias (the latter corresponds to a sin-
gle-equation probit model). We find that the probit model for the election of depreciable
PP&E for revaluation is not independent of the selection model for the revaluation
option for PP&E: the null hypothesis of the independence of the two models is rejected
at the one-percent level with a chi-square statistic of 20.08. Thus, we focus on the esti-
mation results of the probit model that controls for the selection bias.
We posit that companies are more likely to select depreciable classes of PP&E for
revaluation when they need to improve their financial position or enhance the borrow-
ing capacity. Consistent with our hypothesis, we find that leverage, equity depletion,
and losses are significant determinants of the propensity to elect depreciable PP&E for
revaluation but the ratio of land to PP&E is not: the estimated coefficient on Land
Intensity is insignificant at the conventional level. Compared to the selection model,
namely, Model (2), in Table 6, ROA, negative operating cash flows (NegCFOD), and
revaluations in the past (PastRevalD) are no longer significant but the incidence of
equity depletion (EqDepletD) is significant at the one-percent level. Companies that do
not meet the minimum legal-capital covenant are more likely to revalue depreciable
assets in addition to land. On balance, our results suggest that companies revalue depre-
ciable classes of PP&E (mostly along with land) when they need to improve their finan-
cial position or enhance the borrowing capacity rather than signal better future
prospects or mitigate political costs.
4.4. Additional analysis for revaluation decrements

Companies can selectively revalue PP&E, but there is a constraint on their discretion
over the choice of items to revalue: companies are required to revalue all items in the
chosen classes of PP&E. This clause intends to mitigate opportunistic behavior on the
part of revaluing companies; that is, without this clause, companies may revalue assets
that have appreciated, but leave out those that have depreciated below the carrying
amount. When companies revalue PP&E mainly to increase their net assets, they will
not be interested in recognizing valuation decrements as these are costly for revaluing
companies, because they decrease the reported earnings on the income statement. We
argue that a revaluation decrement is an unintended consequence rather than an out-
come that revaluing companies set out to attain.
We conjecture that companies are more likely to recognize revaluation decrements

when they revalue depreciable classes of PP&E. The depreciation process in accounting
is not intended to measure a decrease in value, but does reduce the gap between the fair
value and the carrying amount of depreciable assets. In addition, land prices have been
increasing in recent years. So, the possibility of revaluation decrements will be greater
for depreciable classes of PP&E.
We also conjecture that companies are more likely to recognize revaluation decre-
ments when they employ an independent valuer. Companies are not required to hire an
independent valuer to determine the fair value. Considering that management has less
incentive to recognize revaluation decrements and that the determination of the fair
value depends on many estimates and assumptions, valuations made by independent
valuers are likely to be more objective than those made by managers. Thus, the possi-
bility of revaluation decrements will be greater when companies employ an independent
valuer.
We do not expect that companies adopt the revaluation model for PP&E mainly to
recognize revaluation decrements. In fact, even if companies do not adopt the revalua-
tion model for PP&E, the current impairment rules for long-lived assets mandate the
recognition of the impairment of PP&E as a loss on the income statement.
Table 8 presents the univariate and multivariate test results for our conjectures for
revaluation decrements. We focus on the multivariate test results presented in Panel B.
In our sample of the 302 revaluing companies, 78 companies (26%) recognize reval-
uation decrements as well as revaluation increments, while 224 companies (74%) recog-
nize only revaluation increments. As in the probit model for electing depreciable classes
of PP&E for revaluation (Table 7), we control for the selection bias. However, we are
not able to reject the null hypothesis of the independence of the probit model for recog-
nizing revaluation decrements and the selection model for the revaluation option for
PP&E. Hence, we focus on the estimation results of a single probit regression, which
does not control for the selection bias.
We employ three new variables: revaluation of depreciable classes of PP&E, that is,
other than land (RevalDepPPED); use of an independent valuer (IndpepValuerD); and
the amount of the net increase in PP&E from revaluation (NetRevalInc). We predict
that companies that RevalDepPPED are more likely to recognize revaluation decre-
ments. Land is not subject to depreciation and the price of land has been increasing in
recent years. Hence, the possibility of valuation decrements would be higher for depre-
ciable assets than for land. Our estimation result is consistent with this prediction. The
estimated coefficient on RevalDepPPED is significantly positive at the one-percent
level.
Companies are not required to hire an independent valuer to determine the fair val-
ues of PP&E. Revaluations can be undertaken by managers. In our sample, only 17
companies (5.6%) do not employ an independent valuer. We predict that the probability
of recognizing revaluation decrements will be higher when an independent valuer is
employed, because valuations by independent valuers are likely to be more objective
than those made by managers. However, our estimation result does not support this pre-
diction. In fact, the 17 companies revalue only land and use the OALP, which are the
basis for property-related taxes. Since OALP is determined by the government, the valu-
ation of land based on OALP would be as objective as the valuation by an independent
valuer.
We include NetRevalInc to control for the potential association between recognizing
revaluation decrements and the amount of the net increase in PP&E from revaluation.
Table 8. Recognizing revaluation decrements vs. revaluation increments.
Companies
recognizing only
Companies recognizing both revaluation
revaluation decrements and increments
increments (N = 78) (N = 224) Tests of differences in
Mean Median
Variable (expected sign) Mean Median Mean Median (t-statistic) (Z-statistic)
Panel A: univariate tests of differences in means and medians
RevalDepPPED (+) 0.282 0.000 0.138 0.000 (2.55)⁄⁄ (2.87)⁄⁄⁄
IndepValuerD (+) 0.949 1.000 0.942 1.000 (0.23) (0.22)
NetRevalInc (?) 0.539 0.432 0.654 0.452 ( 1.52) ( 0.32)
Leverage (+) 0.729 0.730 0.723 0.716 (0.19) (0.50)
(3.39)⁄⁄⁄ (2.78)⁄⁄⁄
ROA (?) 0.065 0.007 0.191 0.053

Size (+) 19.797 19.418 18.599 18.234 (5.59)⁄⁄⁄ (5.68)⁄⁄⁄
EqDepletD (+) 0.090 0.000 0.116 0.000 ( 0.68) ( 0.64)
NegCFOD (+) 0.513 1.000 0.545 1.000 ( 0.48) ( 0.48)
LossD (+) 0.423 0.000 0.625 1.000 ( 3.11)⁄⁄⁄ ( 3.10)⁄⁄⁄
PastRevalD (+) 0.474 0.000 0.313 0.000 (2.50)⁄⁄ (2.57)⁄⁄
MTB (?) 0.699 0.661 0.860 0.739 ( 0.41) ( 1.05)
⁄⁄⁄ (⁄⁄)
Significant at the one (five) percent level in a two-tailed test.
Coefficient (p-value)
Controlling for the Single probit
Variable Expected sign selection bias regression
Panel B: probit regression for the recognition of revaluation decrements
Intercept ? 5.732 5.604
(<0.01) (<0.01)
RevalDepPPED + 0.811 0.820
(<0.01) (<0.01)
IndepValuerD + 0.296 0.297
(0.40) (0.40)
NetRevalInc ? 0.147 0.150
(0.31) (0.29)
Leverage + 0.311 0.162
(0.83) (0.76)
ROA ? 1.104 1.642
(0.87) (0.72)
Size + 0.261 0.266
(<0.01) (<0.01)
EqDepletD + 0.337 0.316
(0.42) (0.42)
NegCFOD + 0.153 0.135
(0.49) (0.45)
LossD + 0.314 0.331
(0.22) (0.14)
ROA LossD ? 0.428 1.068
(0.95) (0.82)
PastRevalD + 0.264 0.248
(0.25) (0.17)
MTB ? 0.011 0.010
(0.52) (0.53)
(Continued)
Table 8. (Continued).
Coefficient (p-value)
Controlling for the Single probit
Variable Expected sign selection bias regression
Wald chi-square (p-values) 51.51 (<0.01) 51.87 (<0.01)

Test of independence with the 0.01 (0.91) N/A
selection model: chi-square
(p-values)
dred and two companies revalued some classes of PP&E. Independent variables, where applicable, are
measured on a before-revaluation basis. RevalDepPPED is an indicator variable for companies that
elect depreciable classes of PP&E for revaluation. IndepValuerD is an indicator variable for companies
that employ an independent valuer. NetRevalInc is the amount of revaluation increments, net of the
revaluation decrements divided by PP&E. Leverage is the ratio of total liabilities to total assets. ROA
is the return on total assets. Size is the logarithm of total assets. EqDepletD is an indicator variable
equal to one if the rate of legal-capital depletion, defined by ((legal capital total equity)/(legal capi-
tal)), is at least 50%. NegCFOD is an indicator variable for companies with NegCFOD. LossD is an
indicator variable for companies that reported losses. PastRevalD is an indicator variable for companies
with past revaluation experiences. MTB is the market-to-book ratio.
The p-values in parentheses are from a two-tailed test of z-statistics based on the robust standard errors.
Some companies may select a class of PP&E that contains some assets requiring
recognition of revaluation decrements, because the amount of revaluation increments
from the class is bigger than the amount of revaluation decrements. Consistent with this
explanation, in our sample, there is no single case in which revaluation decrements
exceed revaluation increments and many companies report only revaluation increments.
However, we do not find a significant association between the probability of recogniz-
ing revaluation decrements and the amount of the net increase in PP&E from
revaluation.
Among the other control variables, only size is significant at the one-percent level:
large companies are more likely than small companies to recognize revaluation decre-
ments. We may interpret this finding as follows: politically sensitive large companies
are more likely than small companies to recognize revaluation decrements as well as
revaluation increments. Large companies also may have more items in the class chosen
for revaluation so that the probability of revaluation decrements would be greater for
large companies than for small companies.
5. Summary and concluding remarks

In the absence of a tax benefit, Korean companies are more likely to revalue PP&E
when (1) there are more eligible assets for revaluation (more PP&E), (2) the ratio of
property to PP&E is high, (3) financial leverage is high, (4) there is a greater need for
financing, and (5) they revalued fixed assets in the past. Regarding the signaling content
of asset revaluation, there are two perspectives. On the one hand, companies may signal
better future prospects by electing to revalue PP&E, because asset revaluation reduces
future profitability as measured by ROA or ROE. On the other hand, a company may
signal a better financial position by electing not to revalue PP&E. Our empirical results
support the latter view that companies signal better future prospects or strong financial
position by electing not to revalue PP&E or foregoing the potential benefits of
revaluation.
Several studies have examined the determinants of the propensity to revalue PP&E,
but few of them have investigated (i) why some companies elect both land and depre-
ciable classes of PP&E for revaluation when a majority of companies elect only land
for revaluation and (ii) why some companies recognize revaluation decrements as well
as revaluation increments when a majority of companies recognize only revaluation
increments. These observed asymmetries are related to management discretion over the
timing of revaluation and the choice of items to revalue. Our study provides Korean
evidence on these matters. We find that companies are more likely to revalue deprecia-
ble classes of PP&E in addition to land when they are highly levered, experience equity
depletion, and report losses, that is, when they have a greater need for financing. We
also find that companies are more likely to recognize revaluation decrements as well as
revaluation increments when they revalue depreciable classes of PP&E and have a large
asset base. These findings suggest that revaluation decrements may not be an outcome
that companies set out to attain. Rather, they may be a consequence of revaluations
when large companies with many heterogeneous items revalue PP&E or when compa-
nies try to increase the net revalued amount of PP&E by revaluing a class of PP&E that
happens to contain items for which decreases in value are to be recognized.
Korean companies will face another choice of valuation model for PP&E when they
make a formal transition to IFRS. In 2009, two years ahead of the mandatory adoption
of IFRS, several companies adopted IFRS early but none of them elected the revalua-
tion model for PP&E. These early adopters include the three companies that elected the
revaluation model for the 2008 accounting period. All of them switched back to the
cost model upon the adoption of IFRS in 2009, suggesting that asset revaluations are
opportunistic rather than reflective of economic realities. We conjecture that a majority
of the companies that elected the revaluation model in 2008 will switch back to the cost
model when they make a formal transition to IFRS in 2011. We leave this issue for
future research.
Acknowledgements
We thank Seok Woo Jeong, Myung-In Kim, an anonymous reviewer, and participants at the 2009
Korean Accounting Association Winter Meeting and 2010 American Accounting Association
Annual meeting for helpful comments and suggestions. This work was supported by the National
Research Foundation of Korea Grant funded by the Korean Government (NRF-2010-332-
B00200).
Notes
1. Korean companies are known to suffer from the “Korea Discount” partly due to opacity of
their financial reporting and weak corporate governance (Choi et al. 2012). The introduction
of IFRS coupled with the asset revaluation option could allay the Korea Discount.
2. Asset revaluation has a big impact on the financial position. The total assets of listed Korean
companies have increased by 20.6 trillion KRW or 18.7 billion US dollars and the average
financial leverage (measured as the ratio of total liabilities to total assets) has decreased by
0.092.
3. Barlev et al. (2007) include Korean companies in their sample, but their sample period is
from 1993 to 2004 during which either the now revoked Asset Revaluation Act was applica-
ble or no revaluations were allowed.
4. One notable exception is the treatment of the revaluation surplus when an asset is disposed
of. The current standard requires the recognition of the revaluation surplus in earnings when
the asset is derecognized, whereas IAS 16 prohibits the recognition of the revaluation surplus
in earnings but allows a transfer to retained earnings. The Korean rule is similar to the UK
rule prior to 1993 (Black et al. 1998).
5. Under IFRS, the revaluation of intangibles is also permitted.
6. However, Barlev et al. (2007) do not find evidence for companies in countries with the Brit-
ish-American accounting system.
7. Australian firms used to have another option, i.e. to disclose vs. recognize the fair value for
real estates (Cotter and Zimmer 2003). In this paper, we do not consider this possibility,
because it is not allowed under IFRS.
8. All companies with a nonzero revaluation loss have a nonzero balance in the revaluation sur-
plus, indicating that no revaluing companies report only revaluation decrements.
9. There is one company that revalued depreciable assets, but did not revalue land. Exclusion of
the company does not change the tenor of empirical results.
10. Of the 78 companies, 15 companies are identified through the process of hand collecting and
confirming data for revaluation from annual financial statements. Companies often report
revaluation decrements as impairment losses.
11. It is the sum of the figures for the “independent valuer only” and “mixed” categories.
12. In our sample, the revaluation increment is the same as the balance of the revaluation surplus
because it is the first year of the adoption of asset revaluation.
13. We infer the marginal tax rate used by the company by dividing the amount of deferred tax
liability by the sum of the after-tax revaluation increment and the associated deferred tax lia-
bility. Most companies use a tax rate of 22%, which was the corporate income tax rate for
incomes above 200 million KRW in 2009. We also considered an effective tax rate for
income from continuing operations because revaluation decrements are reported as a part of
continuing operations. The empirical results are insensitive to the tax rate used.
Using the inferred deferred tax rate, we can separate out the tax effect of the revaluation dec-
rement. The net increase in equity is estimated as the after-tax revaluation increment less the
after-tax revaluation decrement, that is, the (before-tax) revaluation decrement times (1 – the
inferred deferred tax rate). The net effect of asset revaluation on the liabilities is estimated as
the deferred tax liability that is associated with the revaluation increment (as reported by the
company) less than the tax effect of the revaluation decrement. The net effect on assets or
PP&E equals the sum of the effects on equity and liabilities. The net income will decrease by
the after-tax amount of the revaluation decrement. To summarize, the effects of asset revalua-
tion on assets, liabilities, equity, and the net income are calculated as follows:
• Increase in assets = (after-tax) revaluation increment + deferred tax liability for the revalua-
tion increment – (before-tax) revaluation decrement.
• Increase in liabilities = deferred tax liability for the revaluation increment – [(before-tax)
revaluation decrement deferred tax rate].
• Increase in equity = (after-tax) revaluation increment – [(before-tax) revaluation decrement
(1 – deferred tax rate)].
• Decrease in net income = (before-tax) revaluation decrement (1 – deferred tax rate).In the
above, the deferred tax rate = deferred tax liability for the revaluation increment/(after-tax
revaluation increment + deferred tax liability for the revaluation increment).
14. If we focus on companies that report revaluation decrements, the average decrease in the net
income becomes 1542 million KRW or 6.4% of the amount before revaluation.
15. For companies with a December year-end, revaluations are made in or after January 2009.
Note that the market capitalization is measured at the end of the fiscal period, that is, before
revaluations are undertaken.
16. We identify 128 firms that voluntarily disclosed their plans to revalue PP&E and those that
voluntarily disclosed the results of revaluation. We exclude 37 firms that announced an asset
revaluation along with their announcement of earnings, because it is difficult to separate the
effect of asset revaluations from the effect of an earnings announcement. We further delete a
company in which there were no stock returns during our event window.
17. We select a matching non-revaluing firm that is in the same industry (based on the two digit
SIC) and similar in terms of the probability of PP&E revaluation. That is, we ensure that
matching non-revaluing firms are similar in terms of their incentives to revalue PP&E, but
differ in their choice of revaluation. This analysis is performed for 85 pairs of firms for which
we can find a matching firm in the same industry. See Section 4.2 for models to assess the
propensity to revalue PP&E. We use Model (2) in Table 6.
18. For simplicity, we present changes in the carrying amount of each class of PP&E, net of the
associated accumulated depreciation. Namsun discloses that the company did not recognize a
deferred tax asset for the decrement because there would not be enough future profits against
which the deductible temporary difference (revaluation decrements) could be utilized. How-
ever, Namsun could have reduced the deferred tax liability, which is a tactic adopted by many
companies that report revaluation decrements.
19. The debt-to-equity ratio has decreased from 6.15 to 3.38 as a result of the asset revaluation.
References
Barlev, B., Fried, D., Haddad, J., and Livnat, J., 2007. Reevaluation of revaluations: a cross-coun-
try examination of the motives and effects on future performance. Journal of business finance
& accounting, 34, 1025–1050.
Black, E., Sellers, K., and Manly, T., 1998. Earnings management using asset sales: an interna-
tional study of countries allowing noncurrent asset revaluation. Journal of business finance &
accounting, 25, 1287–1317.
Brown, P., Izan, H., and Loh, A., 1992. Fixed asset revaluations and managerial incentives. ABA-
CUS, 28, 36–57.
Choi, T., Lee, E. and Pae, J., 2012. The equity premium puzzle: empirical evidence for the
“Korea Discount”. Asia-Pacific Journal of Accounting & Economics, DOI:10.1080/
16081625.2012.667378.
Cotter, J. and Richardson, S., 2002. Reliability of asset revaluations: the impact of appraiser inde-
pendence. Review of accounting studies, 7, 435–457.
Cotter, J. and Zimmer, I., 1995. Asset revaluations and assessment of borrowing capacity. ABA-
CUS, 31, 136–151.
Cotter, J. and Zimmer, I., 2003. Disclosure versus recognition: the case of asset revaluation. Asia-
Pacific journal of accounting & economics, 10, 81–99.
Easton, P., Eddey, P., and Harris, T., 1993. An investigation of revaluations of tangible long-lived
assets. Journal of accounting research, 31, 1–37.
Gaeremynck, A. and Veugelers, R., 1999. The revaluation of assets as a signaling device: a theo-
retical and an empirical analysis. Accounting and business research, 29, 123–138.
Greene, W., 2003. Econometric analysis. 5th ed. Upper Saddle River, NJ: Prentice-Hall.
Henderson, S. and Goodwin, J., 1992. The case against asset revaluations. ABACUS, 28, 75–87.
Lin, Y. and Peasnell, K., 2000. Fixed asset revaluation and equity depletion in the UK. Journal
of business finance & accounting, 27, 359–394.
Spence, M., 1973. Job market signalling. Quarterly journal of economics, 87, 355–374.
Watts, R. and Zimmerman, J., 1986. Positive accounting theory. Englewood Cliffs, NJ: Prentice-
Hall.
Whittred, G. and Chan, Y., 1992. Asset revaluations and the mitigation of underinvestment. ABA-
CUS, 28, 58–74.
Appendix: A PP&E revaluation example

We illustrate an asset revaluation by Namsun Aluminum Co., Ltd., which is one of the first adopt-
ers of the revaluation option under the revised accounting standard for PP&E. Since its incorpora-
tion in 1973, Namsun has specialized in aluminum-related products such as windows, doors,
curtain walls, and aluminum industrial fabric. Its shares have been listed on the KRX since 1978.
As of 31 December 2008, Namsun had a market capitalization of 35,782 million KRW and total
assets of 158,798 million KRW.
On 21 January 2009, Namsun announced that it would revalue some PP&E. The revaluation
was undertaken by an independent valuer on 31 December 2008, the effective date of the revalua-
tion. The revaluation led to a net increase in PP&E of 22,206.87 million KRW, which comprised
an increment of 22,740.84 million KRW less a decrement of 533.97 million KRW. As a result,
shareholders’ equity increased by 17,203.89 million KRW, which consisted of a revaluation sur-
plus of 17,737.86 million KRW that was included in the accumulated other comprehensive
income less an asset revaluation loss of 533.97 million KRW that was included in the retained
earnings. Namsun also recognized a deferred tax liability of 5002.98 million KRW, which repre-
sented the tax effect of the revaluation increment.
Below is a summary journal entry to account for the revaluation of PP&E (in millions s of
KRW).18
(Dr.) (Cr.)
Land 7788.74
Buildings, net 6888.58
Plant, net 409.95
Machinery, net 7119.60
Revaluation loss (I/S) 533.97
Deferred tax liability 5002.98
Revaluation surplus 17,737.86
There are several interesting observations to make. First, companies are not required to revalue all
PP&Es. Of the eight classes of PP&E, which are disclosed in the notes, Namsun revalued only four
classes: land, buildings, plant, and machinery. Vehicles, tools, equipment, and construction in progress
were not revalued.

Second, all items in the selected classes of PP&E should be revalued at the same time: a
selective revaluation within a class of PP&E is prohibited. That explains why Namsun reports
both a revaluation increment and a decrement. However, the revaluation decrement (533.97 mil-
lion KRW) is much smaller than the revaluation increment (22,740.84 million KRW). A similar
pattern is observed for all companies in our revaluation sample: no company reports a net
decrease in PP&E from an asset revaluation. Companies seem to opt for the revaluation model
when an asset revaluation gives rise to a net increase in PP&E.
Finally, the impact of asset revaluation on the financial statements is large. As a result of
asset revaluation, the assets, liabilities, and shareholders’ equity of Namsun have increased by
22,741 million KRW, 5003 million KRW, and 17,204 million KRW, respectively. They represent
16, 4, and 90% of the respective amounts prior to asset revaluation. The net loss for the year has
increased by 534 million KRW, which is 4% of the net income before revaluation. The ratio of
the total liabilities to total assets (financial leverage) has decreased from 0.86 to 0.77.19

Choi 2013

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Choi 2013

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [Umeå University Library]

On: 03 October 2013, At: 13:24

Asia-Pacific Journal of Accounting &

Asset revaluations: motives and choice

To link to this article: http://dx.doi.org/10.1080/16081625.2012.719858

PLEASE SCROLL DOWN FOR ARTICLE

Asset revaluations: motives and choice of items to revalue

*Corresponding author. Email: jinhanpae@korea.ac.kr

Ó 2013 City University of Hong Kong and National Taiwan University

2. Accounting for asset revaluation and related research

will result in a decrease in future earnings. In contrast, there is no detrimental deprecia-

2.2. Determinants of the propensity to revalue PP&E

2.2.1. Intensities of PP&E and property

2.2.2. Financial leverage

2.2.3. Signaling future performance

2.2.4. Political costs

2.2.5. Equity depletion

2.2.6. Need for ﬁnancing

2.2.7. Past revaluation

2.2.8. Market-to-book ratio

2.3. Choice of Items to revalue

3. Data and descriptive statistics

Table 1. Descriptions of the sample.

Variables Mean Standard deviation First quartile Median Third quartile

PP&E class Land Buildings Machinery Structures Others

Panel B: classiﬁcation by PP&E classes (no. of revaluing companies = 302)

independent valuer. In our sample, the majority of revaluing companies (94.4%)

3.2. Impact of revaluation

Variables Mean Standard deviation First quartile Median Third quartile

Note: % changes are relative to the amount before revaluation.

Table 3. Market reaction to the announcement of an asset revaluation.

Event window ( 1 to +1) ( 2 to +2) ( 3 to +3)

Event window ( 1 to +1) ( 2 to +2) ( 3 to +3)

(t-statistics) (1.25) (0.29) (0.43)

Table 4. Univariate tests of variables affecting the propensity to revalue PP&E.

LossD (+) 0.573 1.000 0.382 0.000

Table 5. Correlations between determinants of the propensity to revalue (p-values).

Table 6. Probit regression – determinants of the propensity to revalue PP&E.

Model (1) Model (2) Model (3) Model (4)

(<0.01) (<0.01) (<0.01) (<0.01)

PastRevalD + 0.264 0.053 0.262 0.049 0.229 0.054 0.198 0.047

4.2. Determinants of the propensity to revalue PP&E (hypothesis 1)

Table 7. Choice of items to revalue – land vs. depreciable assets.

PastRevalD (+) 0.264 0.000 0.373 0.000 ( 1.60) ( 1.51)

Controlling for the selection bias Single probit regression

Wald chi-square (p-values) 85.6 (<0.01) 32.91 (<0.01)

Revaluations will be more beneﬁcial to loss-making companies than to proﬁtable com-

4.3. Choice of Items to revalue

4.4. Additional analysis for revaluation decrements

We conjecture that companies are more likely to recognize revaluation decrements

Table 8. Recognizing revaluation decrements vs. revaluation increments.

ROA (?) 0.065 0.007 0.191 0.053

Wald chi-square (p-values) 51.51 (<0.01) 51.87 (<0.01)

5. Summary and concluding remarks

Appendix: A PP&E revaluation example

were not revalued.

You might also like