The Knowledge Scorecard



INTANGIBLE ASSETS

Measurement, Drivers, Usefulness

By

Feng Gu and Baruch Lev*

Please note: The various intangibles value metrics discussed here were designed by Baruch Lev who retains exclusive rights to the measures, and has a patent pending for them. The measures should not be used or reproduced without a written permission from Baruch Lev.

April 2001

INTANGIBLE ASSETS

1. How are The Intangibles Metrics Computed?

It is widely accepted that intangible (knowledge or intellectual) assets are the major drivers of corporate value and growth in most economic sectors, but the measurement of these assets has eluded so far managers, accountants, and financial analysts valuing investment projects.

Why measure intangible assets? Evaluating profitability and performance of business enterprise, by say, return on investment, assets or equity (ROA, ROE) is seriously flawed since the value of the firm’s major asset— intangible capital—is missing from the denominator of these indicators. Measures of price relatives (e.g., price-to-book ratio) are similarly misleading, absent the value of intangible assets from accounting book values. Valuations for the purpose of mergers and acquisitions are incomplete without an estimate of intellectual capital. Resource allocation decisions within corporations require values of intangible capital. These and other uses create the need for valuing intangible assets, in practically all economic sectors.

Intangible (knowledge) assets, such as new discoveries (drugs, software products, etc.), brands or unique organizational designs (e.g., Internet-based supply chains) are by and large not traded in organized markets, and the property rights over these assets are not fully secured by the company, except for intellectual properties, such as patents and trademarks. The risk of these assets (e.g., drugs or software programs under development not making it to the market) is generally higher than that of physical assets.[1] Accordingly, many, particularly accountants and corporate executives, are reluctant to recognize intangible, or intellectual capital as assets in financial reports, on par with physical and financial assets. While such attitude concerning balance sheets may be understandable, it does not satisfy the need to seek information about and value of intangible assets.

Some have attempted to gauge the value of intangible assets from the difference between the company’s capital market value and its book value (the balance sheet value of net physical and financial assets). This approach is unsatisfactory because it is based on two flawed assumptions: (a) that there is no mispricing in capital markets (tell this to investors who bought Internet stocks in 1999 and saw them plummet in 2000), and (b) that balance sheet historical values of assets reflect their current values.

The market-minus-book approach to valuing intangibles is also unsatisfactory because it is circulatory. One searches for measures of intangibles value in order to provide new information to managers and investors. What is the use of a measure (market-minus-book) that is derived from what investors already know (market and book values)? There is obviously a need for a different approach to estimating the value of intangible assets.

1. Preliminaries:

Baruch Lev’s methodology for measuring the value of intangible assets is based on the economic concept of “production function,” where the firm’s economic performance is stipulated to be generated by the three major classes of inputs: Physical, financial, and knowledge assets. Thus:

Economic Performance = α(Physical Assets) + β(Financial Assets) + δ(Intangible Assets)

α , β and δ represent the contributions of a unit of asset to the enterprise performance.

A key ingredient in this approach is the definition of an enterprise economic performance as an aggregate of past core earnings (earnings excluding unusual and extraordinary items), and future earnings, or growth potential. A performance measure which is strictly based on past earnings or cash flows, or a modification of earnings (e.g., the various value added measures), misses a major part of what intangible assets are all about—creating future growth (e.g., by investment in R&D, Internet activities, or employee training).

Having thus defined enterprise performance, the next step is the measurement of the performance drivers—the three major asset groups. The values of physical and financial (stocks, bonds, financial instruments) assets are obtained from the firm’s balance sheet and footnotes (with proper adjustments, such as converting accounting historical costs to current values). The derivation of the value of the third performance driver— intangible capital—is, in a sense, the solution to the above production function for the one unknown (intangible capital). This is done by estimating the “normal rates of return” on physical and financial assets—the α and β coefficients in the above production function—and subtracting from the estimated economic performance of the enterprise the contributions of physical and financial assets, namely the normal asset returns multiplied by the values of physical and financial assets. What remains from this subtraction is the contribution of intangible assets to the enterprise performance, which I define as “intangibles-driven earnings.” Capitalizing the expected stream of these earnings yields an estimate of “intangible capital.”

The intangibles value measurement procedure is demonstrated graphically in Figure 1.

INTANGIBLE ASSETS

|Past Earnings | |Future Earnings |

| |+ | |

| |Normalized Earnings | |

| | | |

|Subtract: |Return on Physical Assets | |

| | | |

|Subtract: |Return on | |

| |Financial Assets | |

| | | |

|Equal: |Intangibles-Driven Earnings | |

| | | |

| | | |

|Capitalize: |Intangible Assets | |

2. Specifics:

❖ The measurement procedure outlined in Figure 1 starts with the estimation of annual “normalized earnings,” referred to earlier as the performance of the enterprise, which are based on an average of several (generally 3-5) historical years of reported core earnings (net earnings adjusted for extraordinary and other “one time” items), and same number of expected years earnings. For public companies, I use two alternative approaches to estimate expected earnings: consensus earnings forecasts by financial analysts, and an earnings forecast based on the pattern of the firm’s sales. In firm-specific applications, I use various public and proprietary sources to estimate growth potential. Normalized earnings is thus an annual weighted average of 6-10 earnings numbers, giving a heavier weight to expected earnings.

❖ Based on various economic studies and analyses, I estimate the average contributions of physical and financial assets, the α and β in the production function above. For public rankings of companies (Fortune, CFO magazine), I use after tax rates of 7% for physical assets and 4.5% for financial assets, reflecting economy-wide averages. For company-specific applications, I estimate specific rates of return on assets. These rates will change, of course, with market and company conditions.

I then subtract from normalized earnings (defined above), 7% of the value of physical assets and 4.5% of the value of financial assets.

What remains of normalized earnings after these subtractions is the contribution of intangible assets to the enterprise performance, which I define as “intangibles-driven earnings” (IDE).

❖ Lastly, I forecast the series of intangibles-driven earnings over three future periods (a 3-stage valuation model): Future years 1-5, using financial analysts’ long-term growth forecasts (or a sales-based forecast); years 6-10, linearly converging the forecasts to the long-term growth of the economy—3%; and years 11 to infinity, where IDE are assumed to grow annually by 3%—the expected long-term growth rate of the economy.

❖ The discounted value of expected IDE series, using a discount rate which reflects the above-average riskiness of these earnings, yields the estimated of “intangible assets.”

2. How do They Look?

Tables 1 and 2 present the 1999 intangibles measures computed for the five leading companies in 22 nonfinancial industries, followed by the industry median measures. These data constitute the CFO 2001 ranking. [2]

The metrics include the firms’ intangible capital (noted as knowledge cpital in the tables), as of August 2000; their 1999 intangibles-driven earnings; (noted knowledge earnings) and the new value measure—“market-to-comprehensive value” (third column from right). This measure modifies the well-known market-to-book ratio (market value of corporations divided by their book value—net assets on the balance sheet), by adding to the denominator of the ratio the estimated value of the firm’s intangible capital. Thus, the balance sheet value of physical and financial assets (book value), plus the value of intangibles missing from the balance sheet, comprises the “comprehensive value.”

Table 2 indicates, among other things, that many, so called “old economy” industries, are reach in intangibles: aerospace and defence, food and beverages (particularly brands), home products, industrial, oil and gas, retail. Figure 2, based on about 2000 companies for the period 1990-1999, provides a similar message.

We will see below the results of extensive tests demonstrating the unique usefulness of these measures reflecting intangibles assets.

| | | |

| |The Scope of Intangibles | |

  |Name |Industry |Knowledge Capital 8/31/2000 |Knowledge Earnings 1999 |Change in Knowledge Earnings '99-'98 |Knowledge Capital / Book Value |Market Value / Book Value |Market Value / Comprehensive Value |Market Value 8/31/2000 |Return 8/31/2000 - 2/28/2001 | |HON |HONEYWELL INTL INC COM |Aerospace & Defense |33,839 |2,157 |235 |3.6 |3.3 |0.71 |30,891 |22% | |LMT |LOCKHEED MARTIN CORP COM |Aerospace & Defense |27,358 |1,417 |-333 |4.2 |1.8 |0.34 |11,407 |32% | |BA |BOEING CO COM |Aerospace & Defense |23,447 |1,590 |614 |1.9 |3.8 |1.30 |46,270 |17% | |NOC |NORTHROP GRUMMAN CORP COM |Aerospace & Defense |15,901 |894 |65 |4.5 |1.5 |0.28 |5,440 |21% | |RTN.B |RAYTHEON CO CL B |Aerospace & Defense |8,356 |800 |-595 |0.8 |0.9 |0.50 |9,457 |21% | |DAL |DELTA AIR LINES INC DEL COM |Airlines |10,792 |709 |-15 |2.1 |1.2 |0.38 |6,071 |-15% | |AMR |AMR CORP COM |Airlines |9,230 |425 |-174 |1.4 |0.7 |0.31 |4,920 |1% | |LUV |SOUTHWEST AIRLS CO COM |Airlines |6,668 |374 |68 |2.2 |3.7 |1.17 |11,280 |23% | |U |US AIRWAYS GROUP INC COM |Airlines |3,420 |251 |60 |NM |NM |0.72 |2,280 |21% | |AMGN |AMGEN INC COM |Biotech |20,876 |1,041 |136 |6.0 |22.4 |3.20 |77,958 |-5% | |MEDI |MEDIMMUNE INC COM |Biotech |4,409 |124 |36 |6.1 |24.3 |3.44 |17,651 |-48% | |BGEN |BIOGEN INC COM |Biotech |4,377 |219 |44 |4.4 |10.2 |1.90 |10,229 |4% | |CHIR |CHIRON CORP COM |Biotech |1,508 |80 |17 |0.8 |5.4 |2.95 |9,863 |-13% | |DD |DU PONT E I DE NEMOURS & CO COM |Chemical |49,085 |2,543 |23 |3.7 |3.5 |0.75 |46,779 |-1% | |DOW |DOW CHEM CO COM |Chemical |29,091 |1,844 |748 |3.2 |2.0 |0.47 |17,761 |28% | |PPG |PPG INDS INC COM |Chemical |9,948 |632 |63 |3.1 |2.2 |0.53 |7,045 |28% | |APD |AIR PRODS & CHEMS INC COM |Chemical |6,245 |379 |42 |2.4 |2.9 |0.87 |7,746 |13% | |ROH |ROHM & HAAS CO COM |Chemical |4,656 |280 |-29 |1.3 |1.8 |0.77 |6,356 |29% | |IBM |INTERNATIONAL BUSINESS MACHS COM |Computer Hardware |128,186 |6,597 |212 |6.7 |12.1 |1.58 |232,413 |-24% | |DELL |DELL COMPUTER CORP COM |Computer Hardware |83,519 |2,490 |547 |12.9 |17.5 |1.26 |113,251 |-50% | |HWP |HEWLETT PACKARD CO COM |Computer Hardware |49,857 |2,598 |-340 |3.4 |8.2 |1.85 |119,385 |-52% | |EMC |E M C CORP MASS COM |Computer Hardware |45,958 |1,569 |389 |6.9 |32.2 |4.06 |213,677 |-58% | |SUNW |SUN MICROSYSTEMS INC COM |Computer Hardware |44,560 |1,849 |470 |6.1 |27.7 |3.91 |202,719 |-69% | |MSFT |MICROSOFT CORP COM |Computer Software |188,787 |8,526 |2,406 |4.6 |8.9 |1.60 |368,819 |-15% | |ORCL |ORACLE CORP COM |Computer Software |54,304 |2,314 |904 |8.4 |39.4 |4.19 |254,509 |-58% | |CA |COMPUTER ASSOC INTL INC COM |Computer Software |38,908 |1,782 |279 |5.7 |2.7 |0.41 |18,763 |-2% | |VRTS |VERITAS SOFTWARE CO COM |Computer Software |16,988 |176 |143 |5.3 |15.1 |2.40 |48,465 |-46% | |SEBL |SIEBEL SYS INC COM |Computer Software |6,180 |176 |53 |6.9 |45.6 |5.76 |40,715 |-61% | |AES |AES CORP COM |Electric Utilities |28,486 |691 |197 |7.1 |7.3 |0.90 |29,119 |-15% | |DUK |DUKE ENERGY CORP COM |Electric Utilities |15,380 |934 |211 |1.6 |2.9 |1.10 |27,531 |10% | |SO |SOUTHERN CO COM |Electric Utilities |10,351 |847 |177 |1.1 |2.1 |0.99 |19,418 |6% | |FPL |FPL GROUP INC COM |Electric Utilities |5,385 |391 |67 |0.9 |1.7 |0.85 |9,488 |24% | |D |DOMINION RES INC VA NEW COM |Electric Utilities |3,358 |418 |77 |0.5 |1.8 |1.22 |12,604 |26% | |EMR |EMERSON ELEC CO COM |Electrical |24,717 |1,426 |130 |3.9 |4.5 |0.91 |28,273 |2% | |ROK |ROCKWELL INTL CORP NEW COM |Electrical |9,431 |536 |16 |3.5 |2.8 |0.62 |7,534 |15% | |CBE |COOPER INDS INC COM |Electrical |5,950 |363 |27 |3.3 |1.8 |0.43 |3,292 |25% | |APCC |AMERICAN PWR CONVERSION CORP COM |Electrical |4,311 |199 |32 |4.3 |4.6 |0.87 |4,629 |-49% | |KO |COCA COLA CO COM |Food/Beverages |67,165 |3,484 |394 |7.3 |14.2 |1.71 |130,326 |1% | |PEP |PEPSICO INC COM |Food/Beverages |50,480 |2,334 |67 |7.5 |9.1 |1.08 |61,593 |9% | |HNZ |HEINZ H J CO COM |Food/Beverages |18,565 |1,064 |85 |11.4 |8.1 |0.65 |13,223 |14% | |UN |UNILEVER N V N Y SHS NEW |Food/Beverages |18,390 |1,306 |36 |3.0 |4.4 |1.10 |27,007 |19% | |CPB |CAMPBELL SOUP CO COM |Food/Beverages |13,022 |835 |47 |95.1 |81.3 |0.85 |11,140 |20% | |KMB |KIMBERLY CLARK CORP COM |Forest Products |25,308 |1,579 |201 |4.5 |5.6 |1.02 |31,514 |23% | |IP |INTL PAPER CO COM |Forest Products |11,369 |1,103 |841 |0.9 |1.2 |0.63 |15,361 |20% | |GP |GEORGIA PAC CORP COM GA PAC GRP |Forest Products |8,884 |854 |369 |2.2 |1.1 |0.35 |4,568 |13% | |WY |WEYERHAEUSER CO COM |Forest Products |5,762 |572 |285 |0.8 |1.5 |0.81 |10,322 |18% | |WLL |WILLAMETTE INDS INC COM |Forest Products |1,044 |221 |69 |0.5 |1.5 |1.01 |3,331 |54% | |PG |PROCTER & GAMBLE CO COM |Home Products |63,450 |3,882 |143 |5.2 |6.6 |1.07 |80,719 |15% | |G |GILLETTE CO COM |Home Products |26,145 |1,343 |124 |11.0 |13.3 |1.11 |31,590 |9% | |CL |COLGATE PALMOLIVE CO COM |Home Products |19,296 |1,097 |109 |11.8 |17.8 |1.40 |29,257 |17% | |CLX |CLOROX CO DEL COM |Home Products |8,151 |502 |96 |4.5 |4.7 |0.86 |8,517 |0% | |AVP |AVON PRODS INC COM |Home Products |7,675 |455 |24 |NM |NM |1.27 |9,304 |9% | |TYC |TYCO INTL LTD NEW COM |Industrial |56,184 |2,970 |640 |3.7 |6.3 |1.34 |96,177 |-4% | |UTX |UNITED TECHNOLOGIES CORP COM |Industrial |25,856 |1,564 |438 |3.4 |3.9 |0.87 |29,231 |26% | |CAT |CATERPILLAR INC DEL COM |Industrial |23,132 |1,166 |54 |4.2 |2.3 |0.44 |12,705 |15% | |ITW |ILLINOIS TOOL WKS INC COM |Industrial |15,800 |957 |113 |3.1 |3.3 |0.81 |16,922 |9% | |IR |INGERSOLL-RAND CO COM |Industrial |14,453 |819 |77 |4.5 |2.3 |0.42 |7,340 |-4% | |DIS |DISNEY WALT CO COM DISNEY |Media |53,012 |2,126 |59 |2.2 |3.5 |1.07 |82,396 |-20% | |VIA.B |VIACOM INC CL B |Media |16,759 |646 |188 |0.3 |2.1 |1.55 |102,113 |-26% | |CCU |CLEAR CHANNEL COMMUNICATIONS COM |Media |9,536 |447 |119 |0.9 |2.7 |1.40 |27,518 |-21% | |F |FORD MTR CO DEL COM PAR $0.01 |Motor Vehicles |90,338 |6,685 |1,680 |3.7 |2.1 |0.44 |50,941 |18% | |GM |GENERAL MTRS CORP COM |Motor Vehicles |55,026 |4,257 |282 |1.9 |1.3 |0.46 |38,758 |-25% | |DPH |DELPHI AUTOMOTIVE SYS CORP COM |Motor Vehicles |13,413 |962 |97 |3.8 |2.6 |0.54 |9,205 |-14% | |JCI |JOHNSON CTLS INC COM |Motor Vehicles |8,573 |480 |74 |3.5 |1.9 |0.42 |4,589 |26% | |PCAR |PACCAR INC COM |Motor Vehicles |4,159 |306 |-4 |1.9 |1.5 |0.51 |3,246 |13% | |GCI |GANNETT INC COM |Newspapers |17,733 |1,087 |137 |3.8 |3.2 |0.67 |14,928 |18% | |TRB |TRIBUNE CO NEW COM |Newspapers |10,388 |502 |140 |1.7 |1.7 |0.66 |10,999 |14% | |NYT |NEW YORK TIMES CO CL A |Newspapers |5,619 |336 |44 |4.2 |4.9 |0.95 |6,594 |13% | |KRI |KNIGHT RIDDER INC COM |Newspapers |4,921 |329 |12 |3.0 |2.5 |0.63 |4,127 |10% | |DJ |DOW JONES & CO INC COM |Newspapers |3,562 |210 |10 |6.6 |10.1 |1.33 |5,467 |-1% | |XOM |EXXON MOBIL CORP COM |Oil |114,347 |8,544 |878 |1.7 |4.2 |1.57 |284,382 |0% | |RD |ROYAL DUTCH PETE CO NY REG GLD1.25 |Oil |27,258 |3,818 |585 |0.8 |3.7 |2.10 |131,204 |-5% | |CHV |CHEVRON CORPORATION COM |Oil |24,559 |2,210 |1,026 |1.3 |2.9 |1.27 |55,150 |3% | |P |PHILLIPS PETE CO COM |Oil |8,697 |877 |198 |1.7 |3.1 |1.14 |15,756 |-13% | |UCL |UNOCAL CORP COM |Oil |8,453 |376 |42 |3.4 |3.3 |0.74 |8,106 |7% | |PFE |PFIZER INC COM |Pharaceuticals |128,610 |5,796 |3,017 |8.6 |18.2 |1.90 |273,069 |5% | |MRK |MERCK & CO INC COM |Pharaceuticals |109,217 |6,583 |902 |8.6 |12.6 |1.32 |160,694 |15% | |JNJ |JOHNSON & JOHNSON COM |Pharaceuticals |76,446 |4,336 |699 |4.3 |7.1 |1.35 |127,891 |7% | |BMY |BRISTOL MYERS SQUIBB CO COM |Pharaceuticals |74,002 |4,254 |424 |8.3 |11.7 |1.26 |104,255 |21% | |PHA |PHARMACIA CORP COM |Pharaceuticals |55,373 |2,193 |543 |4.7 |6.5 |1.13 |75,998 |-11% | |LLY |LILLY ELI & CO COM |Pharaceuticals |48,163 |2,641 |328 |8.7 |15.0 |1.54 |82,453 |10% | |WMT |WAL MART STORES INC COM |Retail |81,239 |4,867 |1,167 |2.9 |7.5 |1.94 |211,872 |6% | |S |SEARS ROEBUCK & CO COM |Retail |23,457 |1,421 |115 |3.6 |1.7 |0.36 |10,697 |33% | |TGT |TARGET CORP COM |Retail |15,406 |885 |128 |2.6 |3.5 |0.98 |20,999 |68% | |COST |COSTCO WHSL CORP NEW COM |Retail |6,006 |349 |40 |1.5 |3.8 |1.52 |15,404 |21% | |KSS |KOHLS CORP COM |Retail |5,504 |250 |50 |2.9 |9.8 |2.50 |18,486 |18% | |INTC |INTEL CORP COM |Semiconductors |208,641 |9,502 |2,749 |5.7 |13.7 |2.05 |502,711 |-62% | |AMAT |APPLIED MATLS INC COM |Semiconductors |44,667 |1,858 |1,090 |7.3 |11.4 |1.38 |70,011 |-51% | |TXN |TEXAS INSTRS INC COM |Semiconductors |39,390 |1,860 |1,012 |3.1 |8.7 |2.11 |109,810 |-56% | |BRCM |BROADCOM CORP CL A |Semiconductors |5,704 |137 |38 |6.8 |65.8 |8.48 |55,509 |-80% | |HD |HOME DEPOT INC COM |Specialty Retail |48,849 |2,230 |621 |3.5 |8.0 |1.77 |111,287 |-11% | |LOW |LOWES COS INC COM |Specialty Retail |10,962 |567 |171 |2.1 |3.3 |1.06 |17,154 |25% | |CVS |CVS CORP COM |Specialty Retail |10,320 |512 |84 |2.6 |3.7 |1.02 |14,504 |65% | |WAG |WALGREEN CO COM |Specialty Retail |9,243 |510 |73 |2.3 |8.2 |2.50 |33,231 |35% | |RSH |RADIOSHACK CORP COM |Specialty Retail |4,552 |271 |60 |6.3 |15.2 |2.08 |10,962 |-27% | |VZ |VERIZON COMMUNICATIONS COM |Telecom |114,464 |6,462 |1,277 |3.3 |3.5 |0.80 |118,573 |15% | |SBC |SBC COMMUNICATIONS INC COM |Telecom |113,618 |6,903 |2,730 |4.0 |5.0 |1.00 |141,514 |15% | |T |AT&T CORP COM |Telecom |81,221 |4,851 |-222 |0.7 |1.1 |0.62 |118,288 |-26% | |BLS |BELLSOUTH CORP COM |Telecom |53,812 |3,568 |660 |3.3 |4.3 |1.00 |70,185 |13% | |WCOM |WORLDCOM INC GA NEW COM |Telecom |23,277 |1,772 |30 |0.4 |1.9 |1.35 |104,734 |-54% | |CSCO |CISCO SYS INC COM |Telecom Equipment |162,218 |4,910 |2,434 |6.1 |18.5 |2.60 |489,845 |-65% | |LU |LUCENT TECHNOLOGIES INC COM |Telecom Equipment |62,824 |3,220 |315 |2.4 |5.3 |1.57 |139,633 |-70% | |MOT |MOTOROLA INC COM |Telecom Equipment |26,947 |1,684 |1,016 |1.3 |3.7 |1.62 |78,639 |-58% | |GLW |CORNING INC COM |Telecom Equipment |24,786 |867 |210 |3.3 |12.6 |2.97 |96,184 |-75% | |QCOM |QUALCOMM INC COM |Telecom Equipment |19,317 |672 |192 |3.3 |7.7 |1.78 |44,610 |-8% | | NM – Not Meaningful | | | | | | | | | | |

[pic]

Figure 2

3. What Drives Intangible Capital?

Intangible (intellectual) capital is driven by diverse factors: innovation, human capital, organizational processes, customer and supplier relations, to name some major ones. For most of these drivers (e.g., customer satisfaction), there are no standardized, public information available. I, therefore, limit the analysis here to the several intangibles drivers which are publicly available: R&D, advertising (brand support), capital expenditures, information systems, technology acquisition.

Table 3, based on data for about 2000 companies, spanning the period 1989-1999, identifies five major drivers of intangibles-driven earnings (IDE): R&D, advertising (brand enhancement), capital expenditure (intangibles embedded in physical assets), information technology, and technology acquisitions. It is clear from the table that these are indeed drivers-their intensity is positively correlated with the ratio of IDE to sales.[3]

In current work (conducted with Towers Perrin and Feng Gu of Boston University), we find that various measures reflecting human resource practices (e.g., extent of incentive-based compensation, termed LPCT in Table 4, employee training, etc.), are also strongly correlated with intangibles earnings and capital. This is reflected in Table 4.

This is just the beginning of a detailed identification and quantification of the drivers of intangible capital, and in turn, corporate value. Business and investment decisions are predicated on the understanding and quantification of the major drivers of corporate value and growth.

Table 3

Table 4

4. Do They Work?

Given the proliferation of new measures and indicators, proposed to managers and investors, it’s incumbent on the proponents of such measures not only to argue that they are needed and useful, but to prove empirically that they indeed are doing the job. Such a proof is unfortunately missing from most of the proposed measures and analytical techniques.

Below, are comprehensive statistical tests indicating the superiority of the intangibles metrics as indicators of enterprise performance over conventional ones. A frequently used methodology in finance and accounting research to gauge relevance of information and data is to correlate the proposed information with the consequences of investors’ decisions, such as reflected in stock price changes. A weak correlation indicates that the decision makers (e.g., investors) did not find the information very useful, and vice versa for a strong correlation.

Following this approach, I correlated (with Feng Gu) annual stock returns (stock price changes adjusted for dividends), reflecting investors’ decisions, with the annual growth in firms’ intangibles-driven earnings, over the period 1989-1999 (about 2,000 companies in each year). For comparison purposes, I did the same for the annual growth in reported cash flows (from operations) and earnings, two of the most widely used corporate performance measures.

Figure 3 shows the clear superiority of intangibles-driven earnings (IDE), over accounting earnings and cash flows. Specifically, while the correlations between stock returns and reported cash flows or earnings are 0.11 and 0.29, respectively (so much for “cash is king”), the correlations between returns and IDE (based on sales’ growth) is 0.40, and between returns and IDE (based on analysts’ forecasts) is 0.53. Thus, both version of intangibles-driven earnings, with and without analysts’ forecasts, beat earnings and cash flows in the “return correlation race.”

The conclusion: the earnings stream generated by intangible assets (IDE) provide substantially more relevant information to investors than reported earnings and cash flows. The reason: while total earnings, or cash flows reflect the performance of all assets, some of which (e.g., various kinds of physical assets) don’t contribute to growth, IDE focuses on the contribution of intangibles—the major growth contributes. Also, while earnings and cash flows are strictly historic (backward-looking) measures, IDE explicitly reflect growth expectations.[4]

While I cannot perform similar statistical analyses on managerial decisions, analogous to the capital market analysis reported here, it stands to reason that the intangibles metrics reported here will also provide new and useful information for corporate managers. The reason: most corporate decisions are guided by accounting metrics, such as earnings and return on investment measure, which appear inferior to the intangibles metrics.

Figure 3

Table 5

5. Can They Predict?

The statistical validation tests reported in Section 4 were contemporaneous; namely stock return correlated with same year growth in intangibles-driven earnings. Contemporaneous correlations indicate the relevance of an information item to investors. But if the item is widely available, despite it being relevant, you will not be able to use it to gain superior investment returns (the information is already priced).

To test whether intangibles measures can be used to gain “abnormal returns” one must use a multiperiod predictive test. Such a preliminary test is reported in Table 6. With Marc Bothwell of Credit Swiss Asset Management, I estimated for each of the 105 companies in Table 1, its market-to-comprehensive value (M/C) indicator for August 31, 2000. (Recall that the M/C ratio is a modified market-to-book ratio, where the value of intangible capital is added to the denominator). We then correlated the M/C values with the subsequent stock performance of these companies (during September 1, 2000 through December 31, 2000; a period of sharp stock price declines).[5] We found a strong negative correlation, confirming that companies with above-average M/C values (i.e., overvalued by investors, according to the intangibles measures) were subsequently downgraded by investors, and vice versa for companies with below-average M/C value (undervalued companies).

Table 6 indicates that the 53 companies with below-median M/C values (undervalued) gained, on average, 7% in the subsequent period, while the 52 stocks with above-median M/C (overvalued) lost, on average, 15.5% during September-December 2000.

The market-to-comprehensive measure thus appears to distinguish between undervalued and overvalued stocks. With Feng Gu (Boston University) I derive even stronger results for a much longer period:1989-1999, and a larger sample of about 2,000 companies. The M/C investment scheme is profitable during the three-years after portfolio formation and easily beats the widely used measure of Market-to-Book.

Table 7 provides portfolio returns for three investment strategies: book-to-market (B/M), comprehensive-to-market based on analysts’ forecasts (C/M), and comprehensive-to-market based on a sales growth model (AC/M).[6] In each case, the sample companies (about 2000 companies, over 1989-1999) are classified into five portfolios according to increasing size of B/M or C/M. The portfolio return data for one, two, and three years subsequent to portfolio formation indicate: (a) For each year and portfolio strategy, returns are monotonically increasing from the first to the fifth portfolios, a finding documented in finance literature for the B/M portfolios. (b) The increases are steeper for the C/M than for B/M portfolios, see the right column of “Q5 – Q1 Difference” in the three panels of Table 7. (c) The total returns are also higher for the C/M strategy than for the B/M strategy (e.g., for portfolio Q5, the 36 months return is 71.8% for C/M vs. 62.1% for B/M). (d) There are no distinguishable differences in performance between the two versions of C/M; with and without analysts’ forecasts. This is graphically indicated by Figure 4.

Tables 8-10 pit directly the B/M strategy against the C/M portfolio choice. In a series of 5x5 classifications, five by B/M and five by C/M, for 12 months ahead (Table 8) and 24 and 36 months subsequent to portfolio formation (Tables 9 and 10), one can observe the generation of returns for one strategy, when the other is held constant (movement across rows or columns). It is clear from each of the three tables that the significant returns are exhibited across rows, from low to high C/M portfolios (see returns on the right column: CM Q5-Q1). Once the C/M portfolios are accounted for, the B/M portfolio strategy does not generate substantial returns (bottom row in Tables). Thus, the C/M strategy subsumes the well known B/M (“Value”) strategy. Results for AC/M—the intangibles-based comprehensive value based on sales growth model (in contrast with the C/M which is based on analysts’ forecasts), are essentially identical to those using analysts’ forecasts presented in Tables 8-10.

Finally, Tables 11-12 present tests of C/M (or AC/M) portfolio returns adjusted for various risk factors: beta, size, book-to-market, and the “return momentum.” This is the well-known 4-factor model in finance research. The numbers in the tables are risk-adjusted monthly return. It is clear that for both C/M and AC/M, the portfolio returns are sharply increasing from low C/M (AC/M) to high C/M (AC/M) portfolios. The abnormal returns are economically very meaningful. For example, the monthly return for portfolio Q4, 0.236 (Table 11), translate to an annual return of over 3.0 percent above risk benchmark.

Summarizing, the extensive, large sample empirical tests reported in this section indicate that the market-to-comprehensive value metric, based either on analysts’ forecasts or on a sales growth model, exhibit a consistent ability to generate subsequent abnormal stock returns, whether evaluated against a market-to-book strategy, or a combination of risk factors.

[pic]

Table 7

Figure 4

Table 8

Table 9

Table 10

Table 11

Table 12

6. Takeaway Points

❖ The Intangibles Scoreboard adds an essential, and hitherto missing, valuation tool for managers and investors concerned with intangible (intellectual) assets, and with the optimal resource allocation of intangible and physical assets.

❖ R&D, advertising, information technology and various human resource practices were empirically identified as drivers of intangible capital, and in turn corporate value.

❖ Intangibles measures provide more relevant information than conventional performance measures, as indicated by the strength of correlations with stock returns.

❖ Intangibles measures successfully distinguish between over-and under-valued stocks, as indicated by the research presented above.

❖ Lastly, the data and findings reported above are based on publicly

available information, and uniform return and discount rates. It can be expected that substantially improved valuations will be obtained by tailoring the intangibles measures to the specific circumstances of companies, subsidiaries, or stocks.

-----------------------

* Boston University and New York University, respectively

[1] See, Baruch lev, Intangibles: Management, Measurement and Reporting, forthcoming from Brookings Institution Press, June 2001, for elaboration on the unique attributes of intangibles.

[2] This work was done in cooperation with Marc Bothwell, vice president and portfolio manager at Credit Suisse Asset Management.

[3] This work is conducted with Feng Gu of Boston University.

[4] For those interested in a regression analysis, supplementing the univariate correlations of Figure 3, Table 5 provides the appropriate estimates, where annual stock returns are regressed on reported earnings (level and change) and various configurations of the intangibles metrics.

[5] These numbers appear in the right, and third from right columns in Table 1.

[6] In empirical work, the inverse of the multiples (e.g., book-to-market) is preferred , to avoid negative values in the denominator.

-----------------------

Figure 1

Table 6

Table 1

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download