Q&A CASE = HARNISCHFEGER Version: 3/13/10

Harnischfeger Corporation, a large New York Stock Exchange company, faced a financial crisis
in 1982. New management was appointed to turn the company around. As part of its
restructuring strategy, the new management team made a number of financial reporting policy
changes in fiscal 1984. The case describes the company's financial crisis, the turnaround strategy
of the new management team, and the accounting policy changes.

The goals of this classic HBS case are to take the position of Peter Roberts—the decision maker
doing the FSA—and determine whether Harnischfeger has ‗truly‘ turned around, and if so (or if
not) whether Roberts should issue a ―Buy‖ recommendation to his clients. Roberts, by virtue of
being on the equity valuation side of Harnischfeger‘s stakeholders, will care most about future
earnings, cash flows, and default risk over the next three-to-five years (the likely horizon over
which his clients would hold the stock if he recommended a BUY to them).

Specifically, the case requires you to [1] identify whether and where earnings management may
be taking place, and [2] figure out why it is or might be taking place.

Is Harnischfeger experiencing a true turnaround, or is the $1.28 a share it reported in 1984

(improving on losses of $7.64 a share and $3.49 a share in 1982 and 1983) mere financial
smoke? Or is it both (mutually reinforcing?) Why did EPS go up in 1984 but CFOPS down?

CFPS

82 83 84

EPS

BACKGROUND
Harnischfeger is an old, old-line, low-tech, family-based Midwest company. It is fixed-asset
intensive. It has been around for almost 100 years. The Chairman of the Board is Henry
Harnischfeger. It most likely came to be in the situation it is in during 1981-1984 because its
cost structure was fixed, not variable. So when the recession hit in 1980-81, its revenues plunged
and it was unable to reduce its costs quickly or sufficiently. It therefore violated bond covenants

1
left and right, creating financial distress for itself. Henry Harnischfeger has hired a COO
(Goessel) and a new CFO (Grade—a real ―sharpie‖ it turns out!) to fix things.

Q1. Identify up to eight items or areas where it could be argued that Harnischfeger
managed its financial reporting numbers (e.g., but not limited to, earnings) in 1984.
Estimate, as accurately as possible, the effect of each on Harnischfeger’s 1984
reported profits, if any.

Here are arguably eight items or areas where Harnischfeger may have managed its earnings—that
is, purposefully intervened in its financial reporting process—from a recognition point of view.
Per the general definition of ‗earnings management,‘ not all the items relate to net income per se,
and some are more debatable than others. In 1984:-

1a. Harnischfeger retroactively changed its depreciation method from accelerated to straight-line
for all depreciable assets. The cumulative effect of this change, not including the reduction in
the current year's depreciation expense, increased after-tax net income for 1984 by $11.005
million. Until very recently, GAAP said that this kind of retroactive change had to be
reported entirely in the net income of the change year—the firm could not restate prior years
per se. However, the firm did have to provide pro forma indications on its income statement
about what EPS would have been in the two years prior to the change year.1
1b. Harnischfeger did not report any significant reduction in the depreciation expense in 1984
due to this change. That is, the vast bulk of the effect of the change was from ―fixing‖ prior
years, not fixing 1984 per se. (See Note 2.)

2. Harnischfeger changed its estimated depreciation lives of certain U.S. plants, machinery, and
equipment, and the estimated residual values of certain machinery and equipment effective
the beginning of the fiscal year 1984. (See Note 2.) This change increased the pretax
reported profit in 1984 by $3.2 million. This kind of change has to be accounted for
prospectively only—that is, Harnischfeger is not allowed to collect up the lower depreciation
charges that would have arisen had it always had the longer lives and higher estimated
residual values that it is now arguing for. It can only ―capture‖ the benefit of the lower
expense in the year of the change and going forward. No income tax effect was applied to the
$3.2 million in 1984, suggesting that the tax rate to use when converting other pre-tax
amounts created by managing U.S. earnings is zero. This is confirmed in Note 12.

3. During 1984, Harnischfeger changed its rate of return assumption for determining pension
expense. The rate assumed was 9% in 1984 compared to 8% in 1983 and 7.25% in 1982.
During the year, the company also restructured its pension plan and recaptured $39.3 million

1This accounting changed in May 2005 when FASB introduced SFAS 154. This new rule requires that
changes in depreciation method like Harnischfeger’s in 1984 must now be accounted for “retrospectively”
such that the cumulative effect is taken as an adjustment in the beginning of year retained earnings
balance (viz., not as an adjustment of the end of year retained earnings balance via adjusted net income
during the change year).

2
 in excess plan assets. The effect of the change in the rate of return assumption for the
pension plan + the plan restructuring reduced pension expense by approximately $4.00
million in 1984, and this would recur going forward for about ten years. (See Note 11.)

4. The liquidation of LIFO inventories by Harnischfeger increased net income by $2.4 million.
(See Note 7). However, this also increased taxable income by the same amount. Actual
taxes may not have increased because of a zero tax rate in 1984. (See Note 12). Cash flows
would be higher by any tax but also lower because of spending less on inventory (since the
LIFO dip means that they spent less on building inventory than otherwise).

5. Harnischfeger‘s allowance for doubtful accounts as a percentage of gross accounts receivable

fell from 9.1% in 1983 (= $6.4/{$6.4 + $63.74}) to 6.3% in 1984 (=$5.9/{$5.9 + $87.65}).
If the company maintained its allowance for doubtful accounts at 9.1% of its gross accounts
receivable at the end of 1984, its bad debt expense in 1984 would have been {($5.9 + $87.65)
x 0.091} – $5.9 = $2.6 million more than the reported expense. (See Note 8.) Management
has a lot of discretion over how much bad debt expense is, so this is an easy accrual for
management to lever.

A/R (net) Allowance for D/A A/R (gross)

BB $ 63.74 $ 6.4 $ 70.14

9.1%

EB $ 87.65 $ 5.9 $ 93.55

6.3%

6. Effective fiscal 1984, Harnischfeger changed the financial year ending from July 31 to
September 30 for certain foreign subsidiaries. Thus, the 1984 consolidated income statement
included the results of additional months of operations of these subs. This action increased
1984 net sales by $5.4 million. The profit effect of this change was not reported. (See Note
2.) Why didn‘t the foreign subs change their fiscal year to Oct. 31, viz. Harnischfeger‘s fiscal
year-end? It may be that foreign subs take longer to actually close their books so going as far
as Oct. 31 just wasn‘t feasible if Harnischfeger wanted to comply with the SEC‘s US
requirements. Or it may be that Harnischfeger wanted to avoid breaking the three additional
months out in its disclosures, which it may have had to do had the subs moved to an Oct. 31
fiscal year end (since that would be one whole quarter more than 12 months).

7. Harnischfeger‘s R&D expense in 1984 decreased by $7.0 million over the previous year. I
think that most of this reduction was due to Harnischfeger cutting back its R&D in order to
boost EPS. Harnischfeger does have a deal with Kobe in which Kobe is reimbursing
Harnischfeger for $17 million over 1983-85, but the numbers in note 9 are net of those
reimbursements. (See Notes 6 and 9). It would be important for Peter Roberts to determine
if there is a temporary reduction in R&D, or a permanent reduction, or no reduction. If there
was reduction, that might be a bad sign as it would call into question the extent to which

3
 future profits will grow, because in general one needs to invest in R&D to create new
products for the future. However, R&D just may not be economically important for
Harnischfeger relative to what it would be for a high-tech firm.

8. Effective 1984, Harnischfeger began to include in its net sales products purchased from Kobe
Steel, Ltd. and sold to third parties by Harnischfeger. Only the gross margin on Kobe-
originated equipment used to be included in Harnischfeger's financials. This increased
Harnischfeger's sales in 1984 by $28.0 million but had no impact on net income (See Note 2.)
This may have been done to make Harnischfeger‘s ―top-line growth‖ look strong. Many
analysts all but require strong top line growth in order for them to recommend a company.

Although some of the above are pure accounting decisions with no direct cash-flow
consequences, others affect Harnischfeger‘s cash flows. Moreover, some accounting decisions
had one-time impacts while others had multi-year future effects. Figure 1 on p.7 lists the eight
items above and indicates where they lay on those dimensions.

The dimensions of one-time vs. ongoing and whether there were cash flow effects are important
in the business decision facing Peter Roberts because his audience is equity-oriented. They care
about how Harnischfeger‘s stock price performs going forward, and that will depend in large part
on how large and ongoing (i.e., non-transitory) are Harnischfeger‘s future net income/cash flows.

Figure 1 includes more information than asked for by the question because it provides students
an opportunity to take what they‘ve prepared for class and extend it a little in real-time.

A final issue to consider is what the right tax rate is to convert pre-tax impacts into post-tax
impacts. There are three main possibilities:
Use the statutory rate. In 1984, the statutory corporate income tax rate was 46%, and the
average statutory state corporate income tax rate (net of federal income tax benefits) was
about 5%. However, 51% may not the best answer, as note 12 indicates that
Harnischfeger has federal tax operating loss carry-forwards for tax and book purposes.
Use the effective tax rate computed from Harnischfeger‘s 1984 financials, namely $2,425
$5,738 = 42.3%, for each item in points 1-8 above separately.
Use zero. If all of the effects itemized in points 1-8 above pertain to Harnischfeger‘s
Domestic activities, zero is the rate suggested by footnote 12, which indicates that
virtually all the $2.43 million of income tax expense Harnischfeger recognized in 1984
came from its Foreign activities. This is, I believe, the right number.

Figure 1 on p.7 estimates that the after-tax effect of Harnischfeger‘s earnings management on
1984 net income using a zero tax rate was $30.2 million. So had Harnischfeger not done these
changes, its 1984 net income would have been a loss of $15 million, not a profit of $15.2 million,
and one might conclude that Harnischfeger had not experienced much if any of a turnaround.

4
Q2. What do you think are the motives of Harnischfeger’s management in managing its
earnings? Do you think investors will see through what it has done? Explain your
reasoning.

The analysis in Q1 shows that most, if not all, of the reported profits of Harnischfeger in 1984
are produced by accounting changes. Therefore, the accounting changes helped the management
report a significant profit rather than a modest (perhaps even large) loss. Why do did
Harnischfeger‘s management manage earnings via these accounting changes?

There are a number of possible motives. Here are several:

1. Reasons given by management to do with conforming better to industry standards, and

obtaining more accuracy in presenting the situation of the firm.
2. Boost the company's stock price so that the company could raise new capital.
3. Meet the earnings targets of the company's top management compensation plan.
4. Avoid the violation of debt covenant restrictions.
5. Improve the company's image with the customers, dealers, and prospective employees.
6. Changie external financial reporting to help internal people make better pricing and capital
allocation decisions, given that Harnischfeger uses its external financial reporting methods
for input into internal costing and capital budgeting decisions.

Some might argue that the analysis in Q1 shows that it is too complicated for an average investor
to "see through" the impact of all the accounting changes. Others might point out that, even if
many analysts recognize the effect of the company's accounting decisions on the 1984 profits, it
is quite unlikely that the analysts would be able to assess the impact of these changes in future
years. Yet others may argue that the market processes the reported profit numbers efficiently:
there are surely some sophisticated analysts who could perform the analysis in this answer key.

Here is a summary of evidence from research on earnings management & market efficiency:

1. There is considerable evidence in finance and accounting literature that shows that the capital
markets are generally fairly, but not fully, efficient.

2. For stock prices to reflect reality in an unbiased manner, it is not necessary that everyone in
the market has to process the information correctly. There only need be a sufficient number
of sophisticated investors who both "see through" the company's accounting changes and
trade substantial sums of money to back up their views. Is this the case for Harnischfeger?

3. Accounting studies that examine the stock market reaction to accounting changes conclude
that the market does not appear to be permanently fooled by firms‘ accounting decisions,
although it might temporarily fail to understand what‘s going on, and therefore misprice the
stock. However, the evidence presented in these studies is not conclusive. Also, studies do
not examine whether the stock market recognizes the recurring effects of accounting changes.
Without better research, it is difficult to make conclusive statements on this issue.

5
4. Even if capital markets see through the effects of accounting changes, managers may believe
otherwise in making accounting decisions. This is likely to happen if there are no significant
penalties associated with such behavior.

Even if investors fully recognize the impact of Harnischfeger's accounting decisions, there are
other reasons for the company's managers to make these decisions. As the case indicates, the top
management of the company is awarded significant bonuses based on the company's reported
profits (for sure in 1985, and apparently in 1984 (see bottom of p.6). This provides an incentive
for managers to boost short-term profits through accounting changes. However, if the
compensation committee of the company's board of directors recognizes this possibility, the
committee could adjust the reported profits before awarding management bonuses. If investors
can see through these changes from public information, why can't the board do it, especially
when it has access to additional information in the firm?

The third possible motive is the desire of Harnischfeger's management to avoid the violation of
debt covenant restrictions. Since the company recently experienced the painful consequences of
violating these restrictions, it is plausible that the management changed the accounting policies to
avoid future violations of the debt restrictions. If debt covenants are specified in terms of
accounting numbers, managers have an incentive to choose accounting policies to minimize the
violation of the covenants. However, if lenders recognize this possibility, lending agreements
would be modified to avoid it as long as the cost of such a modification is not significant.

The fourth possibility is that the accounting decisions are motivated by a desire to convince the
company's customers, suppliers, dealers, and employees that Harnischfeger is again back on track
and is viable. Given the nature of the company's products, a lack of confidence in the company's
viability is likely to impair the company's ability to sell its products. In fact, the company was
negotiating long-term contracts in 1984 with the governments of Turkey and China. It is quite
possible that the company's return to profitability might have helped the management in this
respect. Similarly, the company's ability to attract and retain talented employees might have been
helped by the image that the company was back on track.

Interestingly, during a visit to the company by Prof. Krishna Palepu, the Harvard professor who
wrote this case, Harnischfeger's management pointed out one additional factor in the company's
accounting decisions: the role of internal management considerations. The company used the
same set of accounting rules for external reporting and for internal management accounting. The
company's product pricing was based on fully allocated product costs, and therefore its
accelerated depreciation policies apparently caused its products to be overpriced relative to
competition. In addition, the higher depreciation charges led to increased capital reinvestment
demands from its divisions for maintaining and replacing the company's fixed assets.

The company's management mentioned three principal reasons for its accounting decisions: (1) a
belief that the external users of accounting data did not adjust for Harnischfeger's conservative
financial reporting when comparing the company's performance with other companies in the
industry, (2) the unpleasant experience with its debt covenant restrictions, and (3) the interaction
between management accounting and external reporting.

6
Underlying all the accounting changes was a reporting philosophy outlined by the then chief
financial officer and the current president of Harnischfeger:

In accounting there is no such thing as absolute truth. The same underlying reality
can be accounted for using a range of assumptions. The earlier philosophy of this com-
pany was to choose the conservative alternative whenever there was a choice. Now we
have decided to change this. We would like to tell the world that we are alive and well.
We wish to tell the truth but do not want to be overly conservative in doing so.
When the outside world compares our financial performance to that of other firms,
they may or may not take the time and effort to untangle the effects of the differences in
financial policies that various companies follow. My own belief is that people adjust
for the obvious things like one-time gains and losses but have difficulty in adjusting for
ongoing differences. In any case, these adjustments impose a cost on the user.
If people adjust for the differences in accounting policies when they compare us
with other companies, then it should not matter whether we follow conservative or
liberal policies. But suppose they do not adjust. Then clearly we are better off
following the more liberal policies than conservative policies. I’m not sure whether
people make the adjustments or not, but either way we wish to present an optimistic
version of the picture and let people figure out what to do with the numbers.
As a company you have to put the best foot forward if you want to raise capital,
convince customers that you are a viable company, and attract talented people to work
for the company. I feel that the financial reporting should help rather than hinder the
implementation of our operating strategy. In my opinion, the changed accounting
format highlights the effectiveness of our strategy better than the old policies do.

Rightly or wrongly, this is the prevailing view among CFOs.

7
 HARNISCHFEGER CASE -- Q1

THIS TABLE SUMMARIZES MY VIEWS ON WHERE, HOW AND BY HOW MUCH HARNISCHFEGER
MAY HAVE ENGAGED IN MANAGING ITS FINANCIAL STATEMENTS (EM) IN 1984

Cash
Actual/possible EM One-
Item Refs. Rev. Exp. Op.Inc Net Inc flow Comments
action in 1984 time?
effects?
Yes for Very visible. BIG effect.
N2 Retroactive change from $11.0, Change made at beginning
1 Depn. 0 + $11.0 No
p.8, 17 Accel. Straight Line No for of 1984 so impact is a
ongoing. cumulative adjustment.
N2 Lengthened estimated
2 Depn. 0 – $3.2 + $3.2 + $3.2 No No Less visible …
p.17 useful lives of PPE
Recaptured excess plan Complex to invert or
N11
3 Pensions assets + expected 0 – $4.0 No understand for typical user
p.21
return on plan assets of financial statements
N7 + $2.4 + $2.4 Small in 1984 relative to
4 Inventory Did LIFO liquidations 0 Maybe ,
p.10, 20 (p.20) (p.10) 1982 and 1983
Took smaller than Tricky calculations to do.
N8 Est.
5 A/R expected provision for 0 Maybe Maybe How much is economic and
p.10, 20 $2.6
bad debts how much is EM?
N2 Changed FYE by 2 mths Transparent but also
6 FYE + $5.4 $ insig. $ insig. Yes No
p.16 for some foreign subs. puzzling (why done?)
N6 + N9 Could be economic reasons
7 R&D Decreased from 1983 0 – $7.0 Maybe
p.10,19-20 here rather than EM
N2 + Makes ‗top line‘ and top
8 GM G.Margin Rev, Exp $ insig. $ insig. No No
p.10, 17 $28.0 line growth look better

Assumes a zero book &

economic tax rate, due to
ACTION
$30.2 Mix Mix large U.S. tax loss carry-
PORTFOLIO
forwards arising from
huge losses in ’82 and ‘83

Version: 1/11/10

7
Q3. Evaluate each component of Harnischfeger's turnaround strategy. Standing in
early 1985, do you think it would have been worthwhile to invest in the company's
stock? Explain why or why not.

Harnischfeger's turnaround strategy has four elements, each of which deserves attention with
regard to deciding whether to invest or not, as of early 1985:

(1) Changes in top management

(2) Cost reductions to lower the company's break-even point
(3) Reorientation of its business
(4) Restructuring its finances to facilitate the implementation of the reorientation strategy.

The changes in the top management seem to be good. The new chief executive officer (CEO)
has considerable experience in Harnischfeger's industry. The new CEO demonstrated his
credibility with the financial community by successfully negotiating with the company's lenders
to restructure the company's debt.

The new management has taken several steps in the right direction. The company's cost-
reduction programs seem to be paying off. These programs were helpful in reducing the
company's losses in 1984.

Harnischfeger's financial management also seems to be sound. The cost-reduction programs and
the pension restructuring have improved the company's cash flow. Harnischfeger has been able
to generate positive cash flow from its operations in 1984. The company raised substantial new
capital through a public offering of debentures and common stock and used the proceeds to pay
off all of the company's restructured debt.

Finally, the company's business strategy seems to be sound. The management recognized the
potential to exploit the company's strength in the material handling equipment business. Through
its Harnischfeger Engineers subsidiary, the company planned to expand in this area and
concentrate on the high margin "systems" business. This strategy is likely to help the company to
move away from the mining and construction equipment business, which is a low-growth and
cyclical industry, to a higher-growth and more stable business.

So…It probably would have been worthwhile to have invested in Harnischfeger‘s stock in early
1985. The worst seemed to be behind the Company, and there were some shrewd accounting,
operating, financing and investing moves being made by senior management.

8
Q4. What happened to Harnischfeger in June 1999?

The following timeline nicely summarizes the key events.

Timeline of Events
May 25, 1999 June 22, 2001
CEO resigns Roof becomes CFO,
Hanson becomes CEO Executive vice president,
Treasurer

June 7, 1999 July 13, 2001

Files Bankruptcy Joy Global Inc
distributes new stock

1999 2000 2001 2002

July 12, 2001

Joy Global Inc.
Emerged from Ch11 Jan 24, 2002
P&H announces
June 1, 1999 layoff of 220 workers
May 18, 2001
Stock plummets
Bankruptcy court
60%
approves reorganization
plan Oct 4, 2001
Senior Notes
Distributed on AMEX

The following events describe the developments subsequent to the time of the case. As can be
seen, Harnischfeger seemed to have succeeded in implementing its strategy effectively. Also, the
company continued to liberalize its financial reporting policies. But eventually things got out of
control because Grade went off the deep end….

1985
1. The company changed its accounting for duration patterns and tooling. Previously, the cost
of the patterns and tooling was expensed in the year of acquisition. Under the new method,
these costs are capitalized and amortized over their estimated useful lives.
2. Harnischfeger reported a net profit of $0.74 per share for fiscal 1985. The accounting change
described above contributed $0.24 per share to the reported profits.
3. The company raised $147 million by issuing preferred stock.

1986
1. Mr. Goessel was appointed as the chairman and CEO of the company, and Mr. Grade was
appointed as the president and chief operating officer (COO). Previously, Mr. Goessel was
the president and COO, and Mr. Grade was the CFO.

9
2. Harnischfeger acquired Beloit Corporation, a producer of papermaking machinery and
systems, for $175 million in cash. Later in the year, stock equivalent to a 20% equity interest
in Beloit was sold to Mitsubishi Heavy Industries, Ltd., for $60 million in cash.
3. The company acquired Syscon Corporation, a firm based in Washington, DC for $92 million
in cash. Syscon developed advanced computer systems for military markets.
4. Harnischfeger announced a plan to sell the company's Construction Equipment Division for
approximately $17 million in cash and $55 million in debentures.
5. The company reported that Harnischfeger Engineers received a major order for the design of
an automated car assembly plant.
6. Harnischfeger reported a net loss of $1.14 per share for fiscal 1986. This consisted of a profit
of $2.15 per share from continuing operations, a loss of $4.45 per share from discontinued
operations (Construction Equipment Division), and a gain of $1.16 per share from the
adoption of the new pension accounting rules.

1987
1. Harnischfeger received a takeover offer from Columbia Ventures, Inc., for $19 per share in
cash. The company considered the offer inadequate and rejected it.

1992
1. In 1992, Goessel retired and Grade was named CEO. Goessel tells the Board that ―You can
either get someone new and train them (to be the CEO), or stick with Jeff (Grade) and his
shortcomings. He will need some control. Someone is needed to monitor him.‖ Grade‘s
colleagues described him as a ―prototypical alpha male‖, very image conscious, hard driving,
with an aggressive can-do attitude and an enormous ego.
2. This worked well for a few years. The stock price skyrocketed from $20 a share in 1992 to
$50 a share in 1997. But it seems that Grade pushed the envelope beyond merely aggressive
but within-GAAP to fraudulent and outside of GAAP when reporting earnings.

1998-1999
1. The result was that the castle crashed. Between 1997 and 1999 the stock price tanks to $10 a
share. Grade is fired. Harnischfeger files Chapter 11 bankruptcy protection.

2001-2010
1. Company emerged from bankruptcy as Joy Global. As the Yahoo! Finance chart on the next
page shows, the firm has done well since then, with the exception of what appears ex-post to
have been the creation and bursting of a small bubble between the 4th quarter of 2005 and 3rd
quarter of 2006. According to Robert Aronen of the Motley Fool (5/30/06), in early May
2006, this was due to high commodity prices making it evident that miners all over the world
were going to be digging more coal, copper, iron ore, and just about everything else out of the
ground. Furthermore, the development of the oil sands in Alberta was creating a new growth
industry for mining equipment manufacturers like Joy Global. But then there was the
unexpected early retirement of Joy Global's CFO, Donald Roof, who departed for the
traditional, ―personal reasons‖, which seemed to bring reality back to the stock price.

10
JOY GLOBAL’s stock price performance since 2001 (as of 1/8/10)

11