ACCOUNTING FOR
INTANGIBLES
By
Baruch Lev
New York University
Tel: 212-998-0028
Email: [email protected]
Web: www.baruch-lev.com
This presentation, or any part of it, should not be reproduced, or
otherwise used in any form without Baruch Lev’s written permission.
© Baruch Lev, April 2003.
All Rights Reserved.
AT A GLANCE
 Intangible Assets are the Major Drivers of
Corporate Value and Growth.
 The Economics of Intangibles: Value-Drivers and
Value-Detractors.
 Intangibles Challenge Investor Valuations.
© Baruch Lev, April 2003. All Rights Reserved.
2
AT A GLANCE - Continued
 The Traditional Accounting System is of Little
Help in Monitoring and Accounting for
Intangibles, and in Providing Useful
Information to Investors.
 Despite Widespread Disbeliefs, Accounting Data
Can be Adjusted, to Better Reflect Value, and
Intangibles Can Be Valued.
© Baruch Lev, April 2003. All Rights Reserved.
3
I. What They Say
Alan Greenspan
From one perspective, the ever-increasing proportion of our GDP that represents
conceptual as distinct from physical value added may actually have lessened cyclical
volatility. In particular, the fact that concepts cannot be held as inventories means a
greater share of GDP is not subject to a type of dynamics that amplifies cyclical
swings. But an economy in which concepts form an important share of valuation has
its own vulnerabilities.
As the recent events surrounding Enron have highlighted, a firm is inherently fragile
if its value added emanates more from conceptual as distinct from physical assets. A
physical asset, whether an office building or an automotive assembly plant, has the
capability of producing goods even if the reputation of the managers of such facilities
falls under a cloud. The rapidity of Enron's decline is an effective illustration of the
vulnerability of a firm whose market value largely rests on capitalized reputation. The
physical assets of such a firm comprise a small proportion of its asset base. Trust and
reputation can vanish overnight. A factory cannot.
Source: Testimony before the Committee on Financial Services, U.S. House of Representatives, February 27,
2002.
4
Leonard Nakamura
I argue that U.S. private gross investment in intangibles is
at least $1 trillion. Most of this investment goes uncounted.
This is not surprising, in that it is new: the rate of investment
in intangibles, and its economic value, accelerated significantly
beginning around 1980.
An intangibles investment rate of $1 trillion suggests that
U.S. businesses are investing nearly as much in intangibles as
they are in plant and equipment.
If we are investing $1 trillion a year in intangibles…then the
long-run equilibrium value of intangibles is $6 trillion.
Source: Leonard Nakamura, " A Trillion Dollars a Year in Intangibles Investment and the New Economy," 2001.
5
ec
-9
M 5
ar
-9
6
Ju
n9
Se 6
p9
D 6
ec
-9
M 6
ar
-9
7
Ju
n9
Se 7
p9
D 7
ec
-9
M 7
ar
-9
8
Ju
n9
Se 8
p9
D 8
ec
-9
M 8
ar
-9
9
Ju
n9
Se 9
p9
D 9
ec
-9
M 9
ar
-0
0
Ju
n0
Se 0
p0
D 0
ec
-0
M 0
ar
-0
1
Ju
n0
Se 1
p0
D 1
ec
-0
M 1
ar
-0
2
Ju
n02
D
Market-To-Book
S&P 500 Average Market-To-Book Ratio: 19952002
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00
Month
Source: Factset
6
II. WHAT ARE INTANGIBLES?
I.
DEFINITION
Intangible assets are sources of future benefits which do not have a physical
embodiment.
II.
THE MAJOR CATEGORIES OF INTANGIBLES
Discovery/Learning:
New products, services, patents, communities of
practice, Intranet, adaptive capacity.
III.
Customers:
Brands, trademarks, on-line distribution
channels,marketing alliances.
Human Resources:
Work and compensation practices, employee
training, incentives.
Organizational Structure:
Business structures and processes, incentive
systems, information and control systems.
ALLIANCES PREVALENT IN DISCOVERY/LEARNING AND CUSTOMERS
INTANGIBLES
© Baruch Lev, April 2003. All Rights Reserved.
7
MANAGERIAL AND INVESTOR
CHALLENGES
Partial Excludability
 Value dissipates
 Hazy property
rights
 Alliance partners
capture IP
Non-Tradeablility
 No exit strategy
 Absence of
comparables
 Absence of
futures markets
© Baruch Lev, April 2003. All Rights Reserved.
High Risks
 Sunk costs
 Technology and
commercial
uncertainty
Information Challenged
 Insufficient
performance data
 Missing inputoutput linkages
 Non-standardized
measures
8
Merck & Co.: Return on Equity (ROE) Based
on Expensing (top curve) and Capitalization
(bottom curve) of R&D
ROE
0.65
0.6
0.55
0.5
0.45
0.4
0.35
0.3
0.25
0.2
Year
0.15
1975
1976
1977
1978
1979
1980
© Baruch Lev, April 2003. All Rights Reserved.
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
9
A MAJOR PROBLEM:
MONITORING ALLIANCES
“Most large companies now have at least 30
alliances, and many have more than 100. Yet
despite the ubiquity of alliances—and the
considerable assets and revenues they often
involve—very few companies systematically track
their performance. Doing so is not a
straightforward task… Our experience suggests
that fewer than one in four has adequate
performance metrics… Few Senior management
teams know whether the alliance portfolio as a
whole really supports corporate strategy.”
Source: "Managing an Alliance Portfolio," MckInsey Quarterly, 2002 Number 3.
10
III. Mismeasurement & Biased
Reporting of Intangibles
 The myth of conservative accounting:
Mismeasurements of Assets and Income of
Intangibles-Intensive Enterprises.
 Endogenous Software Capitalization:
Computer Associates-Yes; Microsoft-No.
 Asymmetric Accounting: Expense Internally-
Generated Intangibles, Capitalize Acquired
Intangibles.
© Baruch Lev, April 2003 All Rights Reserved.
11
Mismeasurement and Biased
Reporting of Intangibles - Continued
 Manipulation with In-Process R&D.
 Manipulation with R&D Expenditures (and
Other Intangibles?).
 Reporting Opaqueness: No Information on
Employee Training, Brand Enhancement,
Software and Technology Acquisitions, R&D
Breakdowns, Innovation Revenues, etc., etc.
 The Myth of Enron, Global Crossing and
Winstar Being Intangibles-Intensive.
© Baruch Lev, April 2003. All Rights Reserved.
12
IV. Overcoming Accounting
Limitations
 Capitalizing Intangibles.
 Forming Investment Strategies From
Intangibles’ Values.
© Baruch Lev, April 2003. All Rights Reserved.
13
Table 1
Descriptive Statistics
Panel A: Distribution of R&D intensity by industry*
Industry (abbreviation; SIC)
Chemicals and Pharmaceutics (Chem.; 28)
Fabricated Metal (Fab.; 34)
Machinery and Computer Hardware (Mach.; 35)
Electrical and Electronics (Elec.; 36)
Transportation Vehicles (Trans.; 37)
Scientific Instruments (Scient.; 38)
Business Services (Bus.; 73)
Total
Obs.
2,981
1,431
3,803
4,311
1,354
3,400
3,223
Mean
0.0423
0.0262
0.0929
0.0836
0.0454
0.0930
0.0659
SD
0.0655
0.0653
0.2081
0.1526
0.0687
0.1405
0.1887
Median
0.0272
0.0075
0.0423
0.0411
0.0203
0.0552
0.0013
20,503
0.072
0.153
0.032
Scient.
119
139
138
130
128
138
138
154
166
172
178
190
183
188
190
192
Bus.
83
106
104
99
98
103
108
115
134
139
171
183
184
192
187
187
Total
792
921
883
839
835
837
828
879
944
967
1044
1102
1105
1137
1125
1103
2543
2193
15341
Panel B: Observations by industry and year (subsequent returns analysis)
Year
Chem.
Fab.
Mach.
Elec.
Trans.
1983
119
79
152
172
68
1984
135
84
178
198
81
1985
126
81
173
185
76
1986
122
73
163
181
71
1987
116
72
169
183
69
1988
116
74
154
188
64
1989
114
71
157
178
62
1990
124
67
172
184
63
1991
135
66
179
197
67
1992
142
67
185
193
69
1993
160
75
190
202
68
1994
170
72
194
223
70
1995
173
68
194
232
71
1996
179
65
197
242
74
1997
180
62
195
242
69
1998
185
61
184
228
66
Total
2296
1137
2836
3228
1108
* R&D Intensity is measured as the ratio of R&D expense to the market value of equity at the end of the fiscal year.
Source: Lev, Nissim, Thomas. 2001. "On the Informational Usefulness of R&D Capitalization."
14
Table 2
Analysis of the Change in the Association of Earnings and Book
Value with Price from Capitalization and Amortizing R&D
Panel A: Statistics from the reported numbers regression (T = 0)
2000
P/A =
β
y 1983
1y
D y + 2 1/A + 3 B0/A + 4 E0/A + 5 DNE  E0/A + 6 DNE + 
Chem.
Fab.
Mach.
Elec.
Trans.
Scient.
Bus.
1/A
0.090
0.330
1.583
9.222
1.071
6.929
0.022
0.132
0.070
0.183
0.440
2.285
0.968
5.399
B0/A
2.056
17.303
0.568
7.546
1.168
16.358
1.130
14.669
0.963
12.540
1.389
14.667
1.309
14.160
E0/A
8.789
15.271
7.341
15.289
6.304
16.410
7.708
20.504
8.661
20.024
4.581
10.350
10.688
19.870
-13.918
-11.630
-8.634
-10.071
-9.976
-7.534
-13.623
-22.056
-21.249
-21.041
-24.621
-18.749
-15.108
-24.083
DNE
0.002
0.017
-0.223
-3.899
-0.063
-1.134
0.054
0.904
0.092
1.561
0.094
1.091
0.037
0.453
R2
0.268
0.418
0.226
0.250
0.458
0.164
0.284
N
2981
1431
3803
4311
1354
3400
3223
DNE  E0/A
Source: Lev, Nissim, Thomas. 2001. "On the Informational Usefulness of R&D Capitalization."
15
Panel B: Percentage increase in R2 relative to T = 0
T
Chem.
Fab.
Mach.
1
2
3
4
5
6
7
8
Elec.
Trans.
Scient.
Bus.
7.950
1.198
4.457
3.983
1.169
11.743
4.087
14.921
19.621
2.175
2.823
7.433
9.408
7.393
9.637
2.175
2.839
18.130
20.441
5.895
6.891
23.392
3.377
10.205
11.730
3.289
20.910
7.656
26.283
28.659
3.803
3.749
9.808
9.186
13.764
15.239
3.531
3.594
20.584
19.638
7.715
7.074
30.590
3.474
8.552
16.185
3.496
18.505
6.004
32.283
3.346
8.123
16.706
3.168
17.351
4.803
Trans.
Scient.
Bus.
2.689
2.902
2.435
2.544
3.309
3.210
2.823
2.615
3.155
2.725
2.583
2.560
2.447
3.171
2.977
2.432
2.288
2.563
2.244
2.116
2.051
1.941
1.949
1.558
Panel C: Test statistic (standard normal) for the change in R2
T
Chem.
Fab.
Mach.
Elec.
1
7.322
1.597
4.223
4.316
2
7.948
1.695
4.272
5.797
3
8.035
1.738
4.247
6.224
4
8.209
1.747
3.960
5.918
5
8.169
1.619
3.471
5.062
6
8.086
1.500
3.061
4.361
7
7.965
1.432
2.651
3.902
8
7.838
1.333
2.294
3.564
Source: Lev, Nissim, Thomas. 2001. "On the Informational Usefulness of R&D Capitalization."
16
Panel D: Statistics from the “optimal” regression
2000
P/A =
β
y 1983
1y
D y +  2 1/A +  3 BT*/A +  4 ET*/A +  5 DNE  ET*/A +  6 DNE + 
Chem.
Fab.
Mach.
Elec.
Trans.
Scient.
Bus.
0.233
0.905
1.541
9.095
1.057
6.933
-0.031
-0.198
0.253
0.673
0.441
2.338
1.111
6.285
B0/A
2.220
24.023
0.643
8.575
1.117
17.525
1.028
16.903
0.970
12.845
1.384
15.761
1.227
14.904
E0/A
10.459
19.549
7.087
15.066
6.743
18.825
8.760
26.113
8.402
20.592
6.918
15.247
11.061
22.588
-13.581
-11.308
-8.671
-10.379
-9.522
-9.356
-13.404
-22.300
-21.049
-22.341
-27.519
-18.699
-18.383
-25.532
DNE
0.480
4.808
-0.218
-3.868
0.030
0.555
0.272
4.645
0.136
2.350
0.367
4.277
0.208
2.578
R2
0.355
0.434
0.249
0.292
0.475
0.198
0.306
N
2981
1431
3803
4311
1354
3400
3223
1/A
DNE  E0/A
Panel E: Mean (first row) and standard deviation (second row) of the “optimal” book value and earnings adjustments
Chem.
Fab.
Mach.
Elec.
Trans.
Scient.
Bus.
BT*/A – B0/A
0.1246
0.1692
0.0189
0.0323
0.0729
0.0847
0.1329
0.1574
0.0408
0.0493
0.0954
0.0868
0.0892
0.1444
ET*/A – E0/A
0.0111
0.0334
0.0002
0.0083
0.0031
0.0232
0.0082
0.0296
0.0026
0.0107
0.0058
0.0332
0.0052
0.0345
P is market value of common equity, A is total assets, D y is a dummy variables that equals one for year y, BT (ET) is pro-forma
book value (earnings) assuming T years useful R&D life (hence T = 0 implies the reported book-value and earnings), and DNE is a
dummy variable that equals one when pre-R&D earnings are negative.
Panel B reports the percentage change in R 2 (relative to the
benchmark T = 0 regression) from using earnings and book value that have been adjusted to reflect R&D capitalization and
amortization over the subsequent T years.
The test statistics in Panel C are calculated as (N .5  mean[r02 – rT2]) / std[r02 – rT2], where
N is the number of observations, r0 is the residual from the benchmark regressions, and r T is the residual from the adjusted earnings
and book value regression.
The test statistics have a limiting standard normal distribution.
T* denotes the value of T that maximizes
the percentage change in R 2 (identified in panel B using bold font).
Source: Lev, Nissim, Thomas. 2001. "On the Informational Usefulness of R&D Capitalization."
17
Cisco Systems
R&D Capitalization and Amortization
($millions)
I. R&D Amortization for 2002 and 2001: Based on Table 2, Cisco’s
R&D life is four years (25% annual amortization).
R&D
Expense
2002
Amortization
2001
Amortization
$3,448
3,922
2,704
2,065
1,614
1,207
-$980
676
516
403
--
--$676
516
403
302
Annual Amortization
2,575
1,897
Original Expense
3,448
3,992
2002
1,893
3,448
(2,575)
2,766
2001
(1,014)
3,922
(1,897)
1,011
Vintage/Year
2002
2001
2000
1999
1998
1997
II. Adjusted Earnings
Reported Earnings
Add: R&D expense
Subtract: R&D amortization
Adjusted earnings
© Baruch Lev, April 2003. All Rights Reserved.
18
III. Adjusted Book Value (equity)
Reported
R&D Vintages:
Current year
Year -1
Year -2
Year -3
Year -4
R&D Capital
Adjusted book value
2002
2001
28,656
27,120
3,448
2,941
1,352
516
0
8,257
3,922
2,028
1,032
403
0
7,385
36,913
34,505
IV. Return on Equity
2002
Reported
Adjusted
© Baruch Lev, April 2003. All Rights Reserved.
1,893
(28,656+27,120)/2
2,766
(36,913+34,505)/2
= 0.068
= 0.077
19
V. Earnings Growth
Reported:
Undefined
Adjusted:
(2,766-1,011)/1,011 = 174%
VI. BVDIST(Book Value Distortion)
1.117x(36,913/46,052-28,656/46,052) = 0.201
© Baruch Lev, April 2003. All Rights Reserved.
20
Figure 2: Cumulative Abnormal Returns (CAR) to BVDIST Portfolios
0 .2
Janua
Janua
0 .1 5
0 .1
Janua
CAR
0 .0 5
0
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
p o rt.
p o rt.
p o rt.
p o rt.
1 (lo w B V D IS T )
2
3
4
p o rt. 5 (h ig h B V D IS T )
-0 .0 5
-0 .1
-0 .1 5
-0 .2
M o n th
Cumulative abnormal return (CAR) is measured as the cumulative sum of the portfolio monthly abnormal returns. Portfolio
monthly abnormal return for each of the 36 months is calculated as the average abnormal return for the corresponding month across
all firm-year observations that “belong” to the portfolio. Monthly abnormal return is calculated as the difference between the firm’s
return and the contemporaneous return on a SIZE and B/M matched portfolio (SIZE and B/M are updated every twelve months).
Source: Lev, Nissim, Thomas. 2001. "On the Informational Usefulness of R&D Capitalization."
21
Baruch Lev, 2002. All rights reserved.
22