| Patent Valuation from a Practical View Point, and | | | | patent value. For example, generally speaking, the |
| Some Interesting Patent Value Statistics from the | | | | broader the claim protection, the more valuable |
| PatentValuePredictor Model - Rick Neifeld, Ph.D., | | | | the patent. |
| Patent Attorney, and President of Neifeld IP Law, | | | | I will tell you that there are good points and bad |
| PC and StockPricePredictor.com, LLC (1) | | | | points about the PatentValuePredictor patent |
| I. Introduction | | | | valuations. First, the valuations are clearly statistical |
| My colleague Grover Rutter (see his article in this | | | | in nature and therefore imperfect. However, there |
| edition) has presented an excellent review of how | | | | is no such thing as perfection in valuation. |
| to treat patents from a financial and tax reporting | | | | Moreover, there is as far as I know, no other |
| standpoint. However, how do you determine the | | | | completely objective and generally applicable |
| real value of a patent? That begs the question: | | | | method of valuing patents. Furthermore, the Web |
| What factors are relevant to the real value of a | | | | implementation of the PatentValuePredictor model |
| patent? You have to know what factors into a | | | | provides immediate results, and it is far less |
| valuation before you can address the valuation | | | | expensive (currently $100 per patent valuation) |
| issue. Let's start with some definitions, and then | | | | than any other method of which I am aware. |
| address this issue. After that, lets look at the | | | | Finally, as the size of an evaluated patent portfolio |
| PatentValuePredictor model and some actual | | | | grows, the PatentValuePredictor model's portfolio |
| valuation data and trends provided by the model. | | | | valuation becomes statistically more accurate. See |
| II. General Valuation Theories | | | | for example the corporate patent portfolio value |
| Valuation is an accounting term which means a | | | | charts in my earlier article "A Macro-Economic |
| lump sum of money payable to receive the | | | | Model Providing Patent Valuation and Patent Based |
| future benefits of an asset at a particular time. | | | | Company Financial Indicators" posted in the |
| There are three generally accepted accounting | | | | publications sections of both and |
| theories for valuing assets: market, cost, and | | | | Finally, there are some other interesting statistics |
| income. Market theory values an asset as the | | | | I would like to share with you that relate to |
| present value ascribed to similar assets in an | | | | valuation of patents. These statistics are derived |
| active public market. Cost theory values an asset | | | | from the PatentValuePredictor model. |
| by the cost of replacing the asset. Income theory | | | | First, there are currently 1,726,307 enforceable |
| values an asset by the present worth of the net | | | | patents. To determine actual dollar values, the |
| anticipated economic benefit of the asset. Can we | | | | PatentValuePredictor model currently assumes |
| apply any of these theories to value patents? | | | | that the entire GDP is covered by patents. The |
| III. Valuation Theories as Applied to Patents | | | | current GDP is $11.252 trillion. Therefore, the |
| Market theory valuation of patents has little or no | | | | PatentValuePredictor model indicates that each |
| utility because no two patents are similar enough | | | | enforceable U.S. patent covers, on average, |
| for the sales price of one to define the value of | | | | annual sales of about $6.5 million (that is, the GDP |
| another. Of course, you can say that patent | | | | divided by the number of currently enforceable |
| licensing and sale applies market theory to reach a | | | | patents). However, profit is, generally speaking, |
| market price. The problem is that most patents | | | | only a small fraction of gross sales, and old |
| are not bought or sold in an arms length | | | | patents near the end of their term have reduced |
| negotiation, and therefore do not have an | | | | value. That explains why the PatentValuePredictor |
| objective sale price. Even when they are licensed | | | | model determines an average value of |
| or sold, the transaction is usually clouded by other | | | | enforceable patents is only about $2.8 million. To |
| factors including tech transfer or line of business | | | | get this result, we calculated the current valuation |
| transfer. | | | | of each one of the 1,726,307 enforceable patents, |
| Cost theory is generally inapplicable since a patent | | | | and then calculated the average value. |
| cannot be replaced. That is, once the invention is | | | | The PatentValuePredictor model indicates that the |
| generally known, it is no longer patentable. | | | | bulk of the most valuable patents are and have |
| Income theorym is applicable in certain | | | | for many years been in the Pharmaceutical or |
| circumstances. Patents have known finite terms. | | | | Biotechnology (Pharma/Bio) technology areas. The |
| If you can determine the income resulting from | | | | chart below shows the currently ten most |
| ownership of a patent over that term, you can | | | | valuable patents and their technology area. |
| assign a value to the patent just like you can | | | | TEN CURRENTLY MOST VALUABLE PATENTS |
| assign a value to a long term bond. | | | | (AS OF 3/11/2004) |
| Conventional methods using income theory to | | | | Patent |
| value a patent analyze micro economic data to | | | | Issued |
| determine the anticipated economic benefit of | | | | Current Value ($) |
| owning the patent. This micro economic data | | | | Assignee |
| includes market data indicating the gross sales and | | | | Technology |
| net income derived from the sale of products | | | | 6,517,866 2/11/2003 1,797,722,689 Pfizer Inc. |
| attributable to the patent, and any revenue | | | | Pharma/Bio |
| derived from licensing the patent. Applying income | | | | 6,500,987 12/31/2002 1,570,968,527 Teva |
| theory to micro economic data to value a patent | | | | Pharmaceutical Industries Ltd. Pharma/Bio |
| is labor intensive, costly, and complex. This | | | | 6,566,344 5/20/2003 1,481,848,538 Idenix |
| method should include an analysis to determine | | | | Pharmaceuticals, Inc. Pharma/Bio |
| the meaning of the claims of the patent, a | | | | 6,465,496 10/15/2002 1,408,931,126 Teva |
| comparison of products to the claims of the | | | | Pharmaceutical Industries, Ltd. Pharma/Bio |
| patent to determine what products are actually | | | | 6,452,054 9/17/2002 1,220,308,695 Teva |
| covered by the patent, a determination of the | | | | Pharmaceutical Industries, Ltd. Pharma/Bio |
| size of the market covered by the patent, and a | | | | 6,221,640 4/24/2001 1,194,927,644 Cubist |
| determination of the cost advantage of the | | | | Pharmaceuticals, Inc. Pharma/Bio |
| patented technology compared to alternative | | | | 6,071,970 6/6/2000 1,107,999,343 NPS |
| technologies for that market. A micro economic | | | | Pharmaceuticals, Inc. Pharma/Bio |
| analysis can be used to prove damages in patent | | | | 6,319,919 11/20/2001 1,081,784,355 Davis; Bonnie |
| infringement litigation. However, a micro economic | | | | (Syosset, NY) Pharma/Bio |
| analysis of a patent is often cost prohibitive for | | | | 5,610,034 3/11/1997 1,071,288,767 Alko Group |
| purposes of business valuation, capital allocation, | | | | Ltd. Pharma/Bio |
| taxes, and licensing. Moreover, the data necessary | | | | 6,022,716 2/8/2000 1,069,310,287 Genset SA |
| for members of the public to perform micro | | | | Pharma/Bio |
| economic analysis of patents is simply not | | | | While the Pharma/Bio tech area has held the lead |
| available. This is because that data includes | | | | for most valuable patents, the relative value of |
| relationships between patents, product lines, | | | | the most valuable patents has been increasing for |
| product line specific costs and earnings | | | | decades. The charts below show the ten most |
| information, and licensing royalty rates and terms. | | | | valuable patents issued respectively in 1983, 1993, |
| Companies rarely release that type of information | | | | and 2003, and a relative measure of their value. |
| to the public. Thus, micro economic analysis of | | | | Note in the sequence of three charts below the |
| patents is often not feasible. | | | | trend of the relative value to increase over the |
| IV. Problems With Generally Applying Income | | | | decades. |
| Theory | | | | TEN MOST VALUABLE PATENTS ISSUED IN |
| I hasten to point out that even income theory | | | | 1983 |
| valuation based upon micro economic analysis has | | | | Patent Issued Relative Value When Issued |
| limited utility in most commercial settings, as | | | | Assignee Technology |
| opposed to its application in patent infringement | | | | 4,399,282 8/16/1983 1,343 Kabushiki Kaisha |
| litigation. Why? Because patents and products do | | | | Yakult Honsha Pharma/Bio |
| not have a one to one relationship. They have a | | | | 4,375,514 3/1/1983 1,256 Schering, |
| many-to-many relationship. As a result, you | | | | Aktiengesellschaft Pharma/Bio |
| cannot simply evaluate the value of a patent once | | | | 4,372,948 2/8/1983 974 Kureha Kagaku Kogyo |
| you know the financials relating to certain | | | | Kabushiki Kaisha Pharma/Bio |
| products that the patent covers. To illustrate this | | | | 4,374,829 2/22/1983 661 Merck & Co., Inc. |
| point, consider the following two hypothetical | | | | Pharma/Bio |
| situations. | | | | 4,396,617 8/2/1983 660 Duphar International B.V. |
| FIRST HYPOTHETICAL SITUATION - UNUSED | | | | Pharma/Bio |
| PATENTS: | | | | 4,399,276 8/16/1983 605 Kabushiki Kaisha Yakult |
| A company owns ten patents. The first patent | | | | Honsha Pharma/Bio |
| covers a first product, and the company | | | | 4,369,189 1/18/1983 551 Union Carbide |
| manufactures that product for a hefty profit. The | | | | Corporation Pharma/Bio |
| second through tenth patents do not cover the | | | | 4,410,537 10/18/1983 507 Burroughts Wellcome |
| first product, but each one covers some | | | | Co. Pharma/Bio |
| alternative potential product that, if produced, | | | | 4,399,148 8/16/1983 499 Union Carbide |
| could effectively compete with the first product. | | | | Corporation Pharma/Bio |
| No one produces anything covered by the second | | | | 4,372,953 2/8/1983 490 Otsuka Pharmaceutical |
| through tenth patent. Do you allocate all value to | | | | Company, Limited Pharma/Bio |
| the first patent? Surely the other nine patents | | | | TEN MOST VALUABLE PATENTS ISSUED IN |
| have actual value to the company! How do you | | | | 1993 |
| allocate income attributable to sales of the first | | | | Patent Issued Relative Value When Issued |
| product to the ten patents in order to assign | | | | Assignee Technology |
| value to each one of the ten patents? | | | | 5,252,474 10/12/1993 1,696 Merck & Co., Inc. |
| SECOND HYPOTHETICAL SITUATION - THE | | | | Pharma/Bio |
| MANY-TO-MANY CONUNDRUM: | | | | 5,256,558 10/26/1993 969 The Trustees of |
| There are three competing products in a particular | | | | Rockefeller University Pharma/Bio |
| niche market and five relevant patents. Patents 1, | | | | 5,258,502 11/2/1993 868 Massachusetts Institute |
| 2 and 3 each cover the first product. Patents 1, 2, | | | | of Technology Pharma/Bio |
| and 4 cover a second product. Patents 2 and 5 | | | | 5,268,273 12/7/1993 824 Phillips Petroleum |
| cover the third product. You also know of the | | | | Company Pharma/Bio |
| existence of prior art that indicates a likelihood | | | | 5,182,263 1/26/1993 823 Hoffmann-La Roche Inc. |
| that some claims in patents 1 and 5 are invalid. As | | | | Pharma/Bio |
| an additional complication, what if it was unclear | | | | 5,187,241 2/16/1993 763 International Business |
| whether certain of the products were in fact | | | | Machines Corporation Pharma/Bio |
| covered by certain ones of the five patents. That | | | | 5,262,568 11/16/1993 756 State of Oregon |
| is, what if the issue of infringement was not cut | | | | Pharma/Bio |
| and dried? Confusing? You bet! Even if you knew | | | | 5,198,563 3/30/1993 695 Phillips Petroleum |
| the sales and profit margins for the various | | | | Company Chem/Polymer |
| products in the hypothetical situations just noted, | | | | 5,227,405 7/13/1993 690 Duke University Pharma |
| there would be no simple or logical way to assign | | | | Bio |
| values to the various patents. This is all too often | | | | 5,196,524 3/23/1993 679 Eli Lilly and Company |
| the reality when comparing patents and products: | | | | Pharma/Bio |
| there exist many-to-many patent-to-product | | | | MOST VALUABLE PATENTS ISSUED IN 2003 |
| relationships of uncertain bounds. | | | | Patent Issued Relative Value When Issued |
| What have I told you so far? I have told you that | | | | Assignee Technology |
| classical approaches to valuation are inadequate. It | | | | 6,517,866 2/11/2003 3,374 Pfizer Inc. Pharma/Bio |
| is time for a new approach. | | | | 6,566,344 5/20/2003 2,646 Idenix |
| V. The PatentValuePredictor Theory for Valuing | | | | Pharmaceuticals, Inc. Pharma/Bio |
| Patents | | | | 6,602,861 8/5/2003 1,252 Research Corporation |
| Now let me tell you about the | | | | Technologies, Inc. Pharma/Bio |
| PatentValuePredictor model for valuing patents. | | | | 6,531,282 3/11/2003 1,225 Oligotrail, LLC Pharma |
| First, you should know that this model is | | | | Bio |
| implemented as a web service, and it provides | | | | 6,605,606 8/12/2003 1,109 Miravant |
| valuations for all U.S. patents and (a provisional | | | | Pharmaceuticals, Inc. Pharma/Bio |
| valuation of) published U.S. patent applications in | | | | 6,665,641 12/16/2003 884 ScanSoft, Inc. |
| real time. | | | | Software |
| How does the PatentValuePredictor model work? | | | | 6,602,503 8/5/2003 861 Biogen, Inc. Pharma/Bio |
| The PatentValuePredictor model simplifies the | | | | 6,596,332 7/22/2003 841 Nestec S.A. Foods |
| valuation determination problem by reformulating | | | | products |
| the problem. It does not attempt to address the | | | | 6,602,499 8/5/2003 824 The General Hospital |
| many-to-many relationship noted above, and it | | | | Corporation Pharma/Bio |
| does not attempt to find and use micro economic | | | | RE038073 4/8/2003 804 Research Corporations |
| data relevant to any particular technology niche. | | | | Technologies, Inc. Pharma/Bio |
| Instead, it substitutes for the foregoing | | | | VI. Conclusion |
| many-to-many quandary and the (generally | | | | Conventional valuation models are not generally |
| unavailable) microeconomic data an estimate of an | | | | applicable to patents. In addition, where they are |
| annual sales covered by the patent. The model | | | | applicable, they are labor intensive, and often |
| generates a nominal annual sales covered by the | | | | require a series of assumptions. The |
| patent based solely upon measurable properties | | | | PatentValuePredictor model is unconventional in |
| of the patent document and the value of the | | | | the sense that it applies a modified version of |
| Gross Domestic Product (GDP). I won't bore you | | | | income theory in which a nominal cash flow is |
| with the details of the model in this article. You | | | | determined based upon a macro economic model |
| can those details of the model in my earlier article | | | | that applies heuristic rules. It has the benefits of |
| entitled "A Macro-Economic Model Providing Patent | | | | being completely objective, applicable to all |
| Based Company Financial Indicators and | | | | patents, inexpensive, and quick. Application of that |
| Automated Patent Valuations " posted in the | | | | model to the universe of U.S. patents indicates |
| publications sections of both and Suffice it to say | | | | that the most valuable patents have generally |
| here that there is a heuristic relationship between | | | | been in the pharmaceutical and biotechnology |
| measurable properties of patent documents and | | | | technology area for the last three decades. |