Innovation, Continuity and Governance: Philips’ survival as industrial giant by financial reporting & disclosure innovations (1912-2004)

Paper to be presented on the twelfth Accounting and Management History Conference Lillie March 29th and 30th 2007

By Dr. Frans G. Volmer Maastricht Univerity (The Netherlands) March, 9 2007

Résumé Innovation, Continuation et Gouvernance: La survivance de Philips comme gérant industriel par les innovations comptables des rapports annuels (1912-2004) La question fondamentale à propos des rapports annuels comptables est de savoir quelle est la contribution des innovations des rapports annuels financière à la gouvernance d’entreprise et la gouvernance à long terme des industries existantes et renommées? Chatfield et Vangermeersch constataient déjà en 1996 que les rapports annuels de Philips SA donnaient une vue d’ensemble passionnante de nombreux problèmes de la comptabilité du vingtième siècle. Pour cette raison, les 92 rapports annuels financiers de Philips de 1912 (année de la fondation de la société anonyme) à 2004 ont été choisis pour l’étude. Cela représente une période de presque un siècle ; ce qui est exceptionnel pour la plupart des recherches comptables sur le court terme. Cette étude établit, en premier lieu, une liste des innovations comptables de Philips : chiffres d’affaires annuels, figures comparatives, comptes consolidés annuels, segmentations, coûts actuels, ressources humaines, flux de trésorerie, profits par action et charge. En second lieu, plusieurs modèles de rapports annuels ont été étudiés pour expliquer l’évolution des rapports annuels durant cette période. En troisième lieu, la question de recherche est: Philips stimulait-t-il systématiquement le financement interne par la non-répartition des profits? Lors de l’étude, aucun élément déterminant n’a été trouvé pour supporter cette hypothèse. Par contre, on a pu constater que les innovations comptables de Philips ont eu lieu dans une période de hauts profits (1950-1954). La dernière question est de savoir si le CEO (P.D.G.) a utilisé des pratiques comptables pour protégerer et consolider son pouvoir. Concernant l’hypothèse « big bath » (faire apparaître les résultats d’une façon qui les rendent pires qu’ils ne le sont) la conclusion est qu’en cas de succession du CEO (PDG) sans transfert de celui-ci vers le Conseil d’Administration, dans 8 des 12 successions, on a trouvé des éléments prouvant cette « big bath » hypothèse. Dans tous les cas , le nouveau CEO (PDG) a commencé avec une perte comptable. La conclusion concernant la période entière c’est que les innovations des rapports annuels peuvent être expliquées par l’hypothèse du financement interne et du nivellement des profits. Lors des années de crise se soldant par de faibles profits, la politique comptable de Philips adoptait une politique comptable tendant à gonfler les bénéfices et pocédait à l’inverse lors des années fastes Mots clés: comptabilité, histoire d’entreprises, innovation comptables, gouvernance d’entreprise

Innovation, Continuity and Governance: Philips’ survival as industrial giant by financial reporting & disclosure innovations (1912-2004)

Abstract The principal research question is “ How financial disclosure innovations contribute to corporate governance and the long term survival of well-known still existing industries?” Chatfield & Vangermeersch stated already in 1996 that the annual reports of Philips' Industries provide a fascinating panoramic view of many accounting issues of the twentieth century. Therefore 92 annual copies of Philips' financial reports since the start in 1912 (as former family business) up to 2004 were selected: nearly a century and that is exceptional in accounting research suffering on short-sighted. In this paper, firstly a list of (Philips') disclosure innovations was established: reporting about sales, comparative figures, consolidated accounts, segmentation, current cost, human resources, cash flows, earnings per share and taxation. Secondly, several financial reporting models were investigated to explain the evolution of Philips' disclosures during the period. Thirdly, the first empirical research question is: Does the financial disclosure policy of Philips stimulate systematically internal financing by retained earnings? No strong evidence was found for this hypothesis. However it was found that Philips' accounting innovations took place in high performance periods (1950-1959). Finally, the second empirical governance research question was investigated: Did the Philips' CEO use creative (big bath) accounting practices as a corporate governance practice in order to protect and consolidate his power? Regarding the big bath accounting hypothesis, it was found that in the case of CEO succession -with no transfer of the former CEO to the Supervisory Board- 8 of the 12 successions provide evidence for big bath accounting. In all this cases the new CEO started with a accounting loss. The conclusion regarding the whole period is that the reporting innovations can be explained with a smoothing/internal financing hypothesis too: in years of crisis with bad performance the new adopted accounting policies tend to increase profits and in booming years the opposite turned out to be true. Keywords: financial accounting, business history, corporate governance, big bath accounting

Contents

2

1.

Introduction

2.

Prior literature about reporting innovations

3.

Financial reporting models and internal financing of the firm

4.

Hypothesis development

5.

Results and discussion

6.

Conclusion

1. Introduction The objective of this paper is to discuss the evolution of Philips’ disclosure practices from 1912 to 2004, placed in the context of accounting history and related to its corporate governance by Philips’ CEO’s. This type of study is valuable considering the importance and size of Philips as an innovative multinational company (consumer electronics) and the fact that disclosure research provides insight in business’ areas of concern. The principal research question is how financial disclosure innovations contribute to corporate governance and the long term survival of well-known still existing industries? Chatfield & Vangermeersch stated in 1996 that “ the annual reports of Philips' Industries provide a panoramic view of many accounting issues of the twentieth century”. The above stated quote provides the ultimate argument to investigate almost a century of accounting disclosures of one of The Netherlands’ most important companies in depth. A very extensive descriptive research in Philips disclosures is provided in the empirical research section. In that part, disclosure size and reporting goals are discussed from 1912 until 2004.. Philips’accounting policies itself are investigated in depth by many researchers such as Vandermeersch (1983), Brink, (1992), Schattke & Vergoossen, (1996), so that these need not be the main focus of this research. The aim is to be able to provide an answer to the question: Was Philips a disclosure innovator and what were its motives? Several points of view are used in this research. Firstly by comparing the timing that individual disclosure items appeared in the annual reports to the timing that they were required by legislation. Second, a set of hypotheses will assess Philips innovativeness, for example based on the three ownership models (Traas, 1976). A comparison between the timing of the average company shift towards the different models and the timing of Philips’ shift is provided. Several advantages regarding agency theory have been identified in prior literature why a company should have a sound disclosure strategy. The third set of hypotheses relates to the above-mentioned ownership models and agency theory. Did Philips stimulated internal financing by retained earnings? Philips needed a lot of money for innovation and through gaining more trust by improved reporting disclosures new investors could be attracted and less dividends had to be distributed. The shift of governance models from proprietary to the closed (entity) model can be explained by the agency theory. Assertions will be made on whether innovations in disclosure are related to a company’s financial performance and whether dividend payouts are related to innovations in disclosure. Expected is that when information increases / improves lower risk premiums (dividends) have to be paid. There are considerable disadvantages to compensation plans that are performance related surrounding CEO turnovers. Taking into account that in the last three CEO successions (which were all non-routine turnovers with no transfer of the former CEO to the Supervisory Board) profits decreased in the transition year, especially the likelihood of big bath accounting seems high. By assessing mutations in the relating balance sheet / profit and loss account items concerning the CEO successions, evidence can be found whether Philips’ CEOs engaged in creative accounting and what were their motives.

3

Royal Philips: A short history1 The principles for what was to become one of the world's biggest electronics companies were laid in Eindhoven, the Netherlands, in 1891. Philips began by making carbon-filament lamps and, by the turn of the century, was one of the largest producers in Europe. As developments in new lighting technologies fueled a steady program of expansion, in 1914 it established a research laboratory to study physical and chemical phenomena and stimulate product innovation. In 1918, it introduced a medical X-ray tube. This marked the beginning of the diversification of its product range and the moment when it began to protect its innovations with patents in areas stretching from X-ray radiation to radio reception. In 1925, Philips became involved in the first experiments in television in 1925 and, in 1927, began producing radios; by 1932, it had sold one million of them. A year later, it produced its 100-millionth radio valve and started production of medical X-ray equipment in the United States. By 1939, when it launched the first Philips electric shaver, the company employed 45,000 people worldwide. Science and technology underwent tremendous development in the 1940s and 1950s, with Philips Research inventing the rotary heads that led to the development of the Philishave electric shaver, and laying down the basis for later ground-breaking work in transistors and integrated circuits. The company also made major contributions to the development of the recording, transmission and reproduction of television pictures. In 1963, it introduced the Compact Audio Cassette. In 1965, it produced its first integrated circuits. The flow of exciting new products and ideas continued throughout the 1970s. Research in lighting contributed to the new PL and SL energy-saving lamps, while Philips Research made key breakthroughs in the processing, storage and transmission of images, sound and data. These led to the inventions of the LaserVision optical disc, the Compact Disc and optical telecommunication systems. In 1972, the company established PolyGram, the enormously successful music recording label. In 1974, it acquired Magnavox and in 1975, Signetic, both in the United States. Acquisitions in the 1980s included GTE Sylvania's television concern and Westinghouse's lamps business. Then, in 1983, came a technological landmark: the launch of the Compact Disc. Other milestones of the time included the production of Philips' 100-millionth TV set in 1984 and, in 1995, the 300-millionth Philishave electric shaver. The 1990s was a decade of significant change for Philips. The company carried out a major restructuring program to return it to a healthy footing, simplifying its structure and reducing the number of business areas. In 1997, in cooperation with several other companies - and building on the success of its Compact Disc technology (invented by Philips and jointly introduced with Sony) - it released what proved to be the fastest growing home electronics product in history: the DVD. Moving into the 21st century, Philips has continued to change and grow. Long aware that for many people it is no more than a consumer electronics producer, it has dedicated itself to projecting a new and more representative image that reflects the products it offers in the areas of Healthcare, Lifestyle and Technology. By following this up in 2004 with a massive advertising campaign to unveil its new brand promise of 'sense and simplicity', the company confirmed its dedication to offering consumers around the world products that are advanced, easy to use and, above all, designed to meet their needs. However, times have changed: in September 2006, Philips sold 80.1% of its Semiconductors business to a consortium of private equity partners. This laid the foundation for a strong and independent new semiconductors company, named NXP, which was founded on the heritage of over 50 years of innovation at the heart of Philips. The sale marks a further milestone in the shift from cyclical activities

1

4

Royal Philips Investor Relations, 2007

to the building of an even stronger company, focused on Healthcare, Lifestyle and Technology, and centered around our brand promise of ‘sense and simplicity’. Or as Fortune2 recently states: Lighting up Philips. CEO Gerard Kleisterlee, Europe businessman of the year, has the Dutch giant glowing again. In one masterstroke, Kleisterlee took a giant step towards accomplishing what his predecessors in the top job has talked about but never done: transform a sleepy European electronics company with mediocre margins into a consumer-oriented powerhouse capable of producing both strong earnings gains and consistent returns for investors (p. 31)

2. Prior literature and theory development Philips history of disclosure and accounting Disclosure research is a topic that has been under considerate attention. As can be read below certain publications are used as a guide to this paper. Zeff, Van der Well and Camfferman’s (1992) ‘Company financial reporting, a historical and comparative study of the Dutch regulatory process contribution’ provides an overview of the Dutch regulatory process and refers to Philips’ accounting practices several times. In Chatfield and Vangermeersch’ ‘The history of accounting, an international encyclopedia a brief overview was given of the accounting history of Philips (1996). Camfferman and Zeff’s ‘Unilever’s financial reporting: 1920’s to the 1940’s’ provide a 20 year overview of Unilver’s disclosure policies. In their paper, even speeches of annual general meetings are used to assess the disclosures. This research however will only be focusing on the annual reports.

Philips’ accounting innovations In this paper references are made to Camfferman’s (1996) ‘Voluntary annual report disclosure by listed Dutch companies, 1945-1983’. This work provides an overview of accounting legislation history of several countries and the special role of the Dutch in this. The Dutch did not have a formal legislation concerning accounting disclosures for a long time. Still, due to the strong accountancy profession the disclosures of the Dutch were at a high level. The special role of Philips is discussed as well. Camfferman’s work identifies 9 disclosure items. In this paper, those 9 items and the 3 models of ownership will be used to assess Philips timing of accounting disclosures placed in the context of tendency of thought and legislation. The nine disclosure innovations are discussed below.

(1) Disclosure of Sales The movement towards disclosure of sales, or turnover, probably illustrated as well as any other change in disclosure the transition from earlier, more secretive, stages in the

5

development of financial reporting to the present, relatively open practices. Sales are an elementary fact of business life, so that already in the early stages of thought on financial reporting it was obvious to all concerned that company managements possessed information regarding sales. On the other hand, disclosures of sales were relatively rare before the Second World War, because fear of competitors were widely perceived as a valid argument for not making this information public. In the later stages of financial reporting regulation, sales became required (Camfferman, 1996, p 172).

(2) Disclosure of comparative figures Information about a particular enterprise increases greatly in usefulness if it can be compared with similar information about other enterprises and with similar information about the same enterprise for some other period or some other point in time. Comparability between enterprises and consistency in the application of methods over time increases the informational value of comparisons of relative economic opportunities or performance. The significance of information, especially quantitative information depends to a great extent on the user’s ability to relate it to some benchmark (Schroeder & Clark, 1995, p 23).

(3) Consolidations When a business organization acquires control over one or more others through the acquisition of a majority of the outstanding voting stock, stockholders of the acquiring company (the parent company) have an interest in the assets of combined parent / subsidiary entity. It is logical to presume that financial statements that combine the results of both parent company and subsidiary operations and financial position would be more meaningful3, at least to parent company stockholders, than presenting the separate financial statements of the parent company and each individual subsidiary company (Schroeder & Clark, 1995, p 604).

2

Fortune, January 29, 2007, page 31-35 In order to be meaningful it is required that the administrations of both concerns are adapted to each other concerning: • Book years • Principals used for valuation and profit determination • Categorization of annual report items 3

6

(4) Disclosure on segments Due to the aggregate form of consolidated financial statements, it is difficult to find out how the various affiliates have performed. The problem is particularly serious when the group is diversified by product line or market. In this case, there is a greater likelihood that the aggregated results mask differences in performance within the group. Companies sometimes help by disclosing segmental information (Sutton, 2000, p 451). Larger diversified companies provide a breakdown of sales and, less frequently in Europe, profit by both industry and geographical area. These disclosures help explain past changes in a firm’s aggregate sales and earnings and make possible more accurate forecasts of its future performances (Sutton, 2000, p 139). Greenstein and Sami found that a downward shift in the relative bid-ask spread is shown to be a function of the number of segments reported (Greenstein & Sami, 1994, p 197). Legislation states that a firm must determine its ‘dominant source and nature’ of its risks and returns. If the dominant source is its lines of business, then the primary segmental report should focus on business segment activities. If it is the geographical areas in which it produces or sells its products, then the important primary segmental should focus on geographical segment activities. The less important segment is the subject of a secondary report. Therefore, the primary report must contain more extensive segmental disclosures than the secondary report (Sutton, 2000, p 670).

(5) Current cost accounting The rationale behind replacement- or current costs4 is that management considers that the accounts provide more up-to-date and therefore more relevant information when assets are stated at current values. The impact of inflation on asset values is especially marked for assets with long or indefinite lives, such as buildings and land. Supporters of the current value methods argue that the income statement5 is more useful, too. Under current value accounting6, the current costs of inputs –and that includes depreciation- are matched against current revenues. Under historical cost accounting, the historical costs of inputs are matched against current revenues (Sutton, 2000, p 218-219).

(6) Employee related disclosure Considerable differences are observed between the disclosures of labor related costs and disclosure of the number of employees. Companies appeared more willing to disclose the number of employees than

4

Replacement cost / current cost refers to the amount of cash or cash-equivalent that would be paid, to acquire an equivalent or the same asset currently, or that would be received to incur the same liability currently (Riahi-Belkaoui, 2004, p 534). 5 Gains / losses are acknowledged by the revaluation reserve. If price fluctuations occur with regard to assets, their value will be adapted towards the price level in the market. If the price rises, the differences between the new level and the old level will not be acknowledged as a profit but, will be absorbed by the revaluation reserve. The revaluation reserve is the amount of equity increase that cannot be paid to investors (Epe & Koetzier, 2002, 242). 6

See Enthoven (1982) 7

their aggregate remuneration, even though they show aspects of the same facet of the enterprise (Camfferman, 1996, p 192). Sutton explains the point of employee related disclosures. Users of financial statements utilize employee data to calculate productivity ratios. They can make use of FTE’s to calculate sales per full time equivalent staff. A better ratio is added value per FTE, which controls for the fact that a company can save on its employee use and have a high turnover. Related to employee numbers obviously is employee cost. High pay and benefits can more than offset high labor productivity. Thus investors and others monitor labor cost ratios as well as productivity ratios (Sutton, 2000, p 514).

(7) Provision of funds statement A major objective of accounting is to present data that allows investors and creditors to predict the amount of cash that will be distributed in the form of dividends and interest, and allow an evaluation of risk. The ability of an enterprise to generate cash from operations is an important indicator of its financial health and the degree of risk associated with investing in the firm (…) the past cash flows from a firm are the best available basis for forecasting future cash flows (Schroeder & Clark, 1995, p 230).

(8) Disclosure of EPS EPS is an important statistic. It’s often cited in the annual report along with sales, operating profits and dividends. Some firms set a minimum annual percentage growth in EPS (= after-tax profit or loss in a period attributable to ordinary (common) shareholders / weighted average ordinary shares outstanding during period) as a financial target. Its influence is greater still because it is employed in other key ratios7.

(9) Taxation related disclosure Under deferred tax accounting, the company records as expense or benefit the tax impact now and in the future, of its activities in the period. Income tax expense has two components: 1.

The income tax currently payable or refundable, known as current tax expense or benefit.

2.

The future tax impact of the current year’s accounting profit or loss, known as deferred tax expense or benefit (Sutton, 2000, p 535-536).

In the Netherlands, changes in tax structure caused deferred tax to occur. Before the Second World War, Dutch companies were not subject to a tax on income or profit. Income was subject to taxation

7

One of these is the price / earnings ratio (= Current market price per ordinary share / most recent 12 months’ EPS). The P/E ratio tells potential investors how many years it will take them to recover, through corporate earnings, an investment made in a company at today’s market price, assuming those earnings remain at current levels. Of course, earnings are not expected to remain at current levels. Investors expect them to grow, the faster the expected rate of growth of earnings, the higher the price investors will pay for them. Thus the P/E ratio is an indicator of a company’s growth prospects, as perceived by the market (Sutton, 2000, p 151). 8

only to the extent that it was distributed as dividends. An important aspect of the switch from a tax on dividends to an income tax was that, without further measures, retained earnings form before the new tax would escape taxation altogether. This was prevented by the transformation of part of retained earnings from previous periods into a tax-free reserve on which the tax authorities retained a deferred claim. Even if, in the future there would be no differences between taxable and reported income, this tax-free reserve confronted companies with the question of whether or not to recognize the deferred claim in their balance sheets (…). Hence in successive stages between 1940 and 1950, Dutch corporations were confronted with, and became more and more aware of, an increasingly complex issue of deferred taxation. The same changes in the tax law that gave rise to the issue of deferred taxation also gave rise to an increased importance of tax costs as a disclosure (Camfferman, 1996, p 182-183).

2.3 Ownership models and disclosure by annual reports Toms (1998) takes an ownership approach to research disclosure in the article: ‘The supply of and demand for accounting information in an unregulated market: examples from the Lancashire cotton mills, 1855-1914’. The conclusions suggest that voluntary disclosure is associated with participatory, democratic ownership structures. Conversely, secretive attitudes are fostered by the centralization of equity ownership around dominating interest groups and by institutionalized systems of collective bargaining. Over the years the purpose of annual reports changed by its changing group of users. Three ownership models that explain the purpose of annual reports are firstly described. After that the agency theory and creative accounting by CEO succession are worked out as explaining tools.

(1) Proprietary model and (2) closed (entity) model In the proprietary model there is no separation between owner and manager. The aim of the company is to enlarge the owner’ capital, only the owner and the tax authorities are direct involved stakeholders. In fact, most of the times these companies only have an annual tax report and no other kind of statements8. In the closed model there is a separation between the owner and the manager. In contrast with the above stated model two parties, except for the tax authorities are interested in information about the company. The use of the annual report is to show the investors how their money was invested and how the company earned its profits. The above stated separation could lead to managers that exploit the gap 8

More specific, the proprietor is the center of accounting interest. Accounting records are kept and statements are prepared from his viewpoint, and are intended to measure and analyze his net worth. Assets represent things owned by the proprietor or benefits accruing to him. Liabilities are his depts. Capital shows the firm’s value to its owner. Revenues immediately increase capital; expenses decrease it. All types of income can be treated very much alike, because all go directly to the owner and increase his wealth (Chatfield & Vangermeersch, 1996, p 480)

9

that exists between the two groups; to their advantage, but to the disadvantage of the shareholders. The solution prescribed has generally been the regular provision of periodic financial information for shareholders (Lee, 1983, p 8). To point one specific year in history that was ‘the start’ of the closed model is a hard and also subjective thing to do. The start of the closed model was in 1950. The following quote from Zeff, et al, will be used as a determining factor of the beginning of the closed model: In the end of the 1940’s, companies were more aware of the importance of the securities market as a source of capital and of the need to provide shareholders with more than the barest essential form one side of the balance sheet (1992, p 80). The annual report is therefore a tool investors use to call management to account.

(3) Towards the Open model Over years, companies became more socialized. This increased the amount of company stakeholders. Not only the investors needed annual reports to generate company information but also more and more parties like, employees, unions, authorities, potential investors and others became interested. Moreover, the annual report became a tool for investors to judge whether they want to invest in the company or not, a tool for banks to judge whether they want to provide loans, etcetera. Due to the increased amount of stakeholder groups, the disclosed information is not solely finance related. This has led to the existence of for example social and environmental reports. The annual report has a function of informativeness and is a decision making tool. After the Second World War, but especially in the 1960’s and 1970’s the socialization of companies started to kick in. From that time on the companies weren’t pictured as being there for the owner’s welfare only. In the Netherlands this development affected legislation. The law on company councils, which enabled employee’s to have influence on the companies activities stems from that time period. In 1970 a new law was accepted, it was an extension of the 1928-1929 Commercial Code (…) and was clearly aimed at greater openness on the part of companies (…) ‘openness’ was broadly interpreted, and included the use of more acceptable principles for valuation and income determination. Specifically the law was aimed at ending the practice of using secret and hidden reserves (Camfferman, 1996, p 73). (…) the minister’s recommendation represented a confirmation of the proposal so strongly endorsed by Kraayenhof at a NIvRA membership meeting. The Nivra made known in its first annual report, issued in the latter part of 1968, that is was setting up an Advisory Committee on Annual Reporting (CAJ, Commissie van Advies inzake Jaarverslaggeving) “to conduct research concerning opinions and insights into the matter of financial reporting of enterprises in The Netherlands and abroad (Zeff et al, 1992, p 188). The initiative taken by the minister eventually led to the formation of the Tripartite Study Group (Zeff et al, 1992, p 190). The private sector Tripartite Study Group was set up in 1970 to make an inventory of the ‘norms of the economic and social climate’ referred to in the 1970 Act; which can be seen as a first step towards the introduction of a system of accounting standard-setting system (Camfferman, 1996, p 83). It was

10

agreed that the preparatory work would be done by the CAJ, and that other interested groups would comment upon its drafts. Thereafter the parties “sitting together” would discuss the drafts (Zeff, et al, 1992, p 193). The fact that three parties are involved in this study group concerning financial statements: auditors, employers and employees makes the atmosphere more open. Additionally, this was the time in which the auditor’s profession was developing. According to Camffermann, it was believed that a growing participation of employees and the general public in corporate affairs required a fundamental reconsideration of the role and the responsibility of the auditor. Issues such as involvement of the auditor in management audits or the possibility of auditors serving as semi-officials on behalf of the public interest began to appear in literature (Camfferman, 1996, p 78). Moreover, the forming of standards and norms can be viewed as an attempt to make financial statements more accessible. The Tripartite Study Group and the year 1970 is the proxy for the starting year of the open model.

2.4 Agency theory and Disclosure (2) The transition towards the closed and open-model has implications for the annual reports. Requirement standards increase by the number of stakeholders. Investors want to have reliable information to make decisions. With regard to the open and closed model, a firm has several objectives to maintain a sound disclosure policy. The objectives will be discussed below. It was found that a sound disclosure policy fosters a healthy relationship between the company and professional financial analysts, and most observers feel that analysts have tremendous influence in today’s investment market (Stanga, 1976, p42). In addition, by providing more information, the investors can assess whether an investment is risky or not. The firm therefore does not have to pay a high premium in order to interest investors in their securities. Thus, firms that practice good disclosure will experience higher stock prices over the long run. In addition to directly benefiting existing stockholders and corporate managers, the higher market prices have a favorable impact on the firm’s cost of capital (Stanga, 1976, p 42). A supporting study of the fact that more information makes it easier to attract capital was one by Greenstein and Sami. They used the bid-ask spread to determine what the impact of a SEC regulation was that brought equal access to information by both uninformed- and informed investors. The findings of this study were limited but drew the expected conclusion that the downward shift in the relative bid-ask spread is shown to be a function of the number of segments reported (Greenstein & Sami, 1994, p 197). Because of the initial less disclosure there was more uncertainty about the company value. In a situation like this the informed traders have greater opportunities to profit. This in turn will cause market makers to widen bid-ask spreads in order to compensate for the increased risk of losing to informed traders. The resultant increase in the bid-ask spread raises the cost of capital (Dechow, Sloan & Sweeney, 1996, p 6)

11

Botosan identifies two streams of research that support the hypopaper that greater disclosure is associated with a lower cost of equity capital. The first stream suggests that greater disclosure enhances stock market liquidity, thereby reducing the cost of equity capital either through reduced transaction costs or increased demand for a firm’s securities. The other stream suggests that greater disclosure reduces the estimation risk associated with investors’ assessments of the parameters of an asset’s return or payoff distribution. These researchers maintain that investors estimate the parameters of a security’s payoff distribution based on available information about the firm (Botosan, 2002, p 23). The conclusion from Botosan’s article is somewhat surprising and provides some nuance to Stanga’s article. Botosan found that greater total disclosure is not associated with a lower cost of equity capital. The results found were that firms that provide greater disclosure in the annual report benefit in terms of a lower cost of equity capital. However, greater other publications disclosure is associated with a higher cost of equity capital. The results provided some support for managers’ claims that greater disclosure of more timely information increases their cost of equity capital, perhaps through greater stock return volatility (Botosan, 2002, p 39). Linked to the above stated, but not so far in to depth, Stanga mentions that good disclosure tends to lessen the fluctuations in a security’s price. The market price will therefore reflect more accurately the firm’s intrinsic value, which increased investor confidence by reducing gambling and speculations (Stanga, 1976, p 43). Next to this, good disclosure eliminates insider profits. The more a firm discloses about its prospects the less superior returns can be achieved by insiders (Stanga, 1976, p 43). A final advantage is that a company with a sound disclosure policy avoids costly requirements that are imposed on the firms by authorized institutions (Stanga, 1976, p 43). Due to accounting scandals, investors currently prefer higher dividend payouts that are more secure than speculating on stock price increases. The predefined innovations will be used as a proxy for accounting quality to find out whether disclosure / accounting quality is negatively related to dividend payouts. Additionally it can be inferred that in order to innovate in accounting / disclosures resources should be available.

2.5 CEO succession, Creative accounting and disclosure (3) Kirschenheiter and Melumad found that reporting a larger earnings surprise reduces the inferred earnings precision, damping the impact on firm value of reporting higher earnings, and providing a natural demand for smoother earnings. They show that for sufficiently ‘bad’ news, the manager underreports earnings by the maximum, preferring to take a ‘big bath’ in the current period in order to report higher future earnings. If the news is ‘good’, the manager smoothes earnings, with the amount of smoothing depending on the level of cash flows observed. Either he over-reports or partially underreports for slightly good news, and gradually increases his underreporting as the news gets better, until

12

he is underreporting the maximum amount for sufficiently good news. This result holds both when investors are ‘naïve’ and ignore management’s ability to manipulate earnings or, ‘sophisticated’ and correctly infer management’s disclosure strategy (2002, p 761). Regarding earnings management, Peek’s article: ‘The use of discretionary provisions in earnings management: evidence from The Netherlands’ argues that between 1989 and 2000, Dutch firms use their discretion in recognizing provisions under two circumstances. To smooth current earnings changes when these are high and to take a big bath when current earnings changes are negative but expected to be transitory (anticipatory big bath accounting) (2004, p 38). Big bath accounting is a strategy of manipulating a company’s income statement to make poor results look even worse9. In Detzler and Machuga’s ‘Earnings management surrounding top executive turnover in Japanese firms’ (2002) a distinction is made between routine and non-routine turnover. In routine occasions horizon and coaching theory can occur. In these instances departing CEO try to smooth earnings during their last years. For non-routine turnovers, weak evidence was found that incoming CEO’s try to depress earnings in their transition year and subsequently increase earnings (big bath) and that departing CEO attempt to cover up weak performance by increasing earnings in their final years Considering all above stated findings and the incentives, CEO’s have10, it is interesting to see whether we can find similarities at Philips. To assess cover up and big bath theory the hypotheses of Detzler and Machuga (2002) are used.

9

Big bath accounting refers generally to the steps taken by management to drastically reduce current earnings per share in order to increase future earnings per share. The situation is similar to a choice of income-decreasing procedures that increase the probability of meeting future earnings’ targets (RiahiBelkaoui, 2004, p 57). 10 The big bath is often implemented in a bad year to enhance artificially next year’s earnings. The big rise in earnings might result in a larger bonus for executives. New CEOs sometimes use the big bath so they can blame the company’s poor performance on the previous CEO and take credit for the next year’s improvements (www.investopedia.com/terms/b/bigbath.asp).

13

3. Financial reporting models and internal financing of the firm 3.1

A short review of the (changes) in Philips’ accounting principles

The main changes in Philips’ accounting principles are stated below: Before 1919

conservative accounting based on historical cost, write-off’s to one guilder, silent reserves, depreciation was treated as a distribution of income

1920-1929 in

a reserve for expansion was created containing money generated by additional paidcapital (1920), in 1924 patents are capitalized

1930-1939

consolidated balance sheet (since 1931), revaluation reserve created, depreciation of capital expenditures charged to reserves

1940-1949

depreciation based on current fixed costs, business cycle equalization reserve introduced for income smoothing, segmental disclosures

1950-1959

introduction current cost accounting (substantialism): capital maintenance in terms of general purchasing power (1950)

1960-1969

excessive technological price falls of inventory is charged to the profit and loss account (1964)

1970-1979

reductions in the value of stocks resulting from technological improvements are charged now to the revaluation account, just as the deferred taxation relating to revaluations

1980-1989

realized revaluation surplus on assets financed by non-equity capital is credited to income (gearing), deferred income taxes over realized revaluations are accounted for as tax expenses, translation exchange differences are charged to shareholders’ interest, provisions for risks of obsolesce and bad debt are fixed at the level of the estimated risk end of the year (the silent reserve is abolished), goodwill is transferred directly to stockholders’ equity, cost of pension plans will be based on future wage trends and the expected rate of return of pension assets

1990-2004

current value accounting is abolished for reasons of simplicity and a move to US GAAP (1992), creation of high restructuring provisions for income smoothing reasons (1990, 1991).

The size of Philips’ financial statements measured in pages rose from 8 pages (1912) to 216 pages (2004). The biggest increases were in the Otten and Kleisterlee periods. The amount of balance sheet items remained relative stable (30/40) over the period, the P & L items grew faster (from 8 to 18). No other Dutch company adopted so many changes in the accounting principals as Philips did. The many changes over time can be explained by Philips’ business surviving policy: reducing book profits and dividends distribution in booming years (1919-1929, 1949-1979) and increasing the profit in crises years (1930-1948, 1980-2004). In the first half of the 20th century Philips used more hidden reserves (write-offs), in the second part more open reserves (revaluations, provisions). Table 1 provides detailed evidence for the period 1986-2005. Statistically no correlation was found between net income and operating cash flows nor between net income and restructuring provisions. However, correlation was found between operating cash flow and restructuring provisions (R square= 0.33 , t=3, P=0.008) and depreciation and capital expenditures (Pierson statistic= 0.534, P=0.05). Finally, Table 1 showed that net income over 20 years (1986-2005) is only 42% of the relating operating cash flow. Therefore, the conclusion is that Philips used a smoothing and internal finance (profit) reporting policy.

14

Table 1 Philips’net income vs cash flow and provisions, and investments vs depreciation (1986-2005) Philips (millions) 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Total

currency

Fl Fl Fl Fl Fl Fl Fl Fl Fl euro euro euro euro euro euro euro euro euro euro euro

net income

959 809 531 792 -4526 981 -900 856 2049 1143 -268 2602 6053 1799 9602 -2475 -3206 695 2836 2868 23200

operating cash flow

capital Depreciation provision expenditure expenses restructuring

2533 3398 365 1203 4256 3414 3631 5734 5029 2502 2638 3210 2140 1913 2996 1248 2228 1992 2697 2090 55217

4667 4756 4145 4133 3537 2956 3330 2576 3541 4899 5021 225 240 286 3170 2143 1161 980 1286 997 54049

3126 3200 3356 3431 3898 3101 3480 3009 2910 1218 1437 1492 1615 1548 1789 1969 1782 1519 1373 1256 46509

6771 6701 6830 6538 9919 8423 9246 8946 9102 3460 3420 3251 2985 3056 3557 3740 3246 2925 2898 2925 107939

3.2 Philips’ accounting innovations The innovations are derived from Camfferman (1996). Below an overview is given of Philips´ timing of including them in their annual report and the timing when the nine innovations became legislation or perceived as desirable by the majority of firms. See paragraph 2, page 5. In the appendix, a table (6) is provided on when a innovation occurred and from which authors the information was retrieved.

(1) Disclosure of Sales In 1976, when most listed companies disclosed their sales, the Tripartite Study Group issued a Considered View that nondisclosure of sales was warranted only when management expected, on objective grounds that substantial disadvantage to the enterprise would ensue from the disclosure (IVb.2.11-12, December 1976) (Camfferman, 1996, p 176). Legislation regarding disclosures of sales can be found in BW 2: 3773a of the Dutch Civil law. Philips started disclosing its sales in 1951 under the leadership of Otten (Annual report, p 9).

(2) Disclosure of comparative figures From Zeff et al. we derive comparative figures were required by the Company Financial Statement Act of 1970 (1992, p 178). Philips started disclosing comparative figures in 1937 (p 12-13), but

15

discontinued due to the Second World War in 1939. In the 1945-1946 annual report the disclosures were picked up again under the leadership of Otten (Philips Annual report, 1945-1946, p 22-23).

(3) Consolidations As explained more fully in Blommaert (1995), the 1970 Act did not contain a strict legal requirement to provide consolidated financial statements. Consolidation was presented as an option next to combined or separate subsidiary financial statements. Of course, for all companies with more than a very small number of subsidiaries, presenting consolidated financial statements would be the only practical way to comply with the legal requirements (Camfferman, 1996, p 201). Consolidation legislation can be viewed in BW2: 406. Due to pressure from shareholders to provide a consolidated balance sheet, the 1931 annual report of Philips already contained a consolidated balance sheet. In the numbers all participations of 50% or more were acknowledged (Philips Annual report, 1931, p 7). In 1948 Philips reintroduced the consolidate balance sheet under the leadership of A.F. Philips.

(4) Disclosure on segments From Camfferman can be derived that since 1978 segmentation played a role in legislation: …however, the adopted version of the directive published in July 1978 restricted the segmentation requirement to sales, which had to be segmented according to both lines of business and geographical area (1996, p 241). Philips started business line segmental disclosures in 1948 (Annual report, p 13) and geographical segmental disclosures in 1952 (Annual report, p 16) expressed in percentages and qualitative information. The business line disclosures were more useful from 1975 on -under the leadership of H.A.C. Riemsdijk- where turnovers and operational results were to be identified (Annual report, p 10). The 1950’s gave birth to geographical segmental disclosures under the leadership of Otten (Philips annual report, 1956, p 12). Legislation regarding disclosures of segments can be found in BW 2: 380 and BW 2: 380 of the Dutch Civil Code.

(5) Current cost accounting Current cost accounting became required in 1980 due to a draft adaptation law to the Fourth Directive. The draft law reflected a number of salient changes from existing Dutch legislation (…): although the governing criterion of ‘considered to be acceptable in the economic and social climate’ in connection with valuation and income determination was retained, current value was, in effect, required to be disclosed in the notes if not actually used in the financial statements (Zeff et al, 1992, p 297). Philips abandoned secret reserve accounting in 1945/1946 and replaced it by a new policy of basing depreciation on the current cost of its fixed assets (Zeff, et al, 1992, p 78-79). Nevertheless 1950 will be the year real replacement value is used. According to Brink replacement value actually used for the first time in 1950 under the leadership of Otten (1992, p 260).

16

(6) Employee related disclosure The 1970 Act, which did have its intended effect on wage cost disclosure, also appears to have had a side effect on the disclosure of the number of employees: even though there was no reference to disclosure of employee numbers, there was an increase of disclosure in the years around 1970 (Camfferman, 1996, p 198). With regard to employee numbers, a disclosure requirement was introduced with the adaptation of Dutch Law to the Fourth Directive in 1983 (Camffermann, 1996, p 192). Philips started disclosing employee numbers in its 1927 annual report under the leadership of A.F. Philips (Annual report, p 6). Wage costs were introduced in the 1952 annual report under the leadership of Otten (Annual Report, p 71). Legislation regarding employee numbers can be found in BW2: 382 and wages in BW2: 3773e&f.

(7) Disclosure of the funds (cash flow) statement 1979 will be used as benchmark to Philips timing of the inclusion of a funds statement: The statement of Source and Application of Funds, on which a mild exposure draft had been issued in September 1979, was the subject of a similarly mild draft Guideline issued in January 1984. One difference was that the 1984 draft recommended that the statement should be included in the annual financial statements as part of the notes (Zeff et al., 1992, p 329). In Philips’ 1971 annual report the first funds statement was included under the leadership of H.A.C. Riemsdijk (Annual report, p 38).

(8) Disclosure of EPS Guidance on earnings per share was published for the first time in December 1976, (Camffermann, 1996, p 234). Philips disclosed EPS from 1961 onwards under the leadership of F. Philips (Annual report, p 3).

(9) Disclosure of taxation From 1955 onwards, disclosure of taxation was required. We derive this from Camfferman: the 1962 Hamburger Committee report, although it largely reiterated the demands of its 1955 predecessor (Rijkens report) was more specific regarding disclosures of tax liabilities. Whereas the Rijkens report had recommended including taxes payable in the ‘other payables’ item, the Hamburger report called for separate disclosure. That deferred taxes were to be separately disclosed might be inferred, but it was not made explicit (1996, p 184). Philips started disclosing its taxes payable separately in the 1954 annual report, under the leadership of CEO Otten, up to that date is was included in the payables item (Philips Annual report, 1954, p 35). From the 1971 annual report the amount of the deferred tax can be identified, in the notes to the consolidated balance sheet (Philips Annual report, 1971, p 35). The 1970 annual report mentions that several items including deferred tax are included in the item ‘overige voorzieningen’ (other reserves) (Philips Annual report, 1970, p 40). Legislation regarding disclosures of taxation can be found in BW2: 375 and BW2: 374.

17

3.3 Philips’ accounting explained by ownership models11 From the tables 6-8 of the appendix can be seen that innovations were numerous in the 1950’s but several innovation already occurred before this decade. Prior to 1950 already 25% of the accounting innovations were in use by Philips, during the 1950’s the innovations where expanded but the first improvements towards investors where made in the previous decade. Additionally current cost accounting had its foundation in 1946 when silent reserves were abandoned (Brink, 1992 p 260). This was the basis for current cost accounting if this innovation was added to the 1940’s, than already 33% of the accounting innovations would be in use before the 1950’s. The innovations had to do with financial matters; Philips became listed12 in New York from 1953-1954 on. The firm therefore became subject to the jurisdiction of the SEC. In the 1950’s the SEC’s policy toward foreign registrants was in its infancy, and evidently, exceptions from the strict regulations applicable to US companies were allowed. Notwithstanding these relaxations in the SEC’s rules, foreign companies knew that they would be competing against US companies for capital13, and they realized that their reporting might be open to suspicion if they did not provide as much disclosure as US companies (Zeff et al, 1992, p 92). Law did, not require all enhancements that were made, yet. Therefore, they are voluntary and an attempt to provide investors with a better insight into the company’s practices. Table 6 in the appendix provide an overview of innovations. During the 1960’s only one of the innovations we identified occurred, the rest after 1970. Therefore, it is necessary to go in depth in the formation of the Tripartite Study Group, the proxy for the closed model. Kraayenhof was one the initiators of the CAJ, Tripartite Study Group and simultaneous, the managing partner of Klynveld, Kraayenhof & Co., which was the auditor of Philips, AKU and Royal Dutch Petroleum, the three most important companies in the country. Kraayenhof maintained a close tough with all three engagements (Interview with G. Timmer, 4 September 1990). Weisglas esteemed Kraayenhof, who had been president of the NIvA in 1944-47, as a ‘big name’ in the auditing profession (Interview with Dr M. Weisglas August 1990) (Zeff et al, 1992, p 96). The CAJ’s founding chairman, A.M. van Rietschoten, senior partner in an audit firm was a former president of the NIvA (…) the other members of the nine-man committee reflected diverse backgrounds, of which P.C. Breek, a member of Philips’ management board was one (Zeff et al, 1992, p 189). From that period on several functionaries of Philips were involved14. 3.4 Disclosures & agency theory

11

See par 2.3 page 9 12 Choi found that firms significantly improve their financial disclosure upon entry into the European capital market (1973, p 170). 13 Cooke found that listing status is an incentive to voluntary disclose more (1991, p 186). Reasons can be to raise capital or to reduce the agency problem. Disclosure in corporate annual reports is one way of reducing shareholders’ monitoring costs (Cooke, 1991, p 177). 14 See also graph 6 in the graph table section of appendix 18

3.4.1 Performance & accounting / disclosure innovations Tables 2a & 2b disclose per decade some important Philips’ key data over the period (1912-2004) Table 2a: Performance & accounting innovations per decade Table 2a a

b

c

d

e

f

g

h

i

1912-1919

466203

8 75% 3 0,24 1 €0,29 1

0

4

0

8

25

0

1920-1929

1488435

6 90% 2 0,17 2 0,22 2 -72% 7

4

0

8

31

1

1930-1939

239079

10 44% 7 0,06 3 0,07 6 -24% 9

4

0

8

47

2

1940-1949

881751

7 60% 5 0,04 4 0,06 7 -52% 8 5% 2 0,72 5

38

0

1950-1959

13749541

3

100 % 1 0,05 4 0,11 3 764% 1 6% 1 0,85 6

19

5

1960-1969

5490741

5 70% 4 0,04 5

43% 5 5% 2 0,80 7

32

1

1970-1979

8349556

4 50% 6 0,02 6 0,07 7

3%

6 2% 3 0,93 4

36

3

1980-1989

47465411

2 70% 4 0,02 6 0,05 8

52% 4 2% 3 1,03 2

29

0

1990-1999

351402812

1 60% 5 0,02 6 0,07 7 115% 3 2% 3 1,14 1

26

0

2000-2004

1365200000

9 60% 5 0,04 5 0,09 5 257% 2 5% 2 0,94 3

31

0

0,1

4

Table 2b accounting innovations per CEO years

a

b

c

d

1910-1921

10

36529

10 70% 4 0,22 1 0,28 1

0

7

29

1922-1938/39

17

143583

9 85% 5 0,09 3 0,11 3 -80% 9

7

37 3

1939/40-1960

21

8382678

7 86% 1 0,06 4 0,11 3 312% 2 4% 2 1,26 1 20 5

1961-1970

10

4946204

8 60% 7 0,04 5 0,09 4

25% 5 4% 2 0,80 8 39 1

1971-1976

6

16562978

6 50% 8 0,02 6 0,07 5

7%

6 3% 3 0,92 7 41 3

1977-1980

4

34714187

11 50% 8 0,02 6 0,05 6

-3%

7 2% 4 0,97 5 47 0

1981-1985

5

51004896

4 80% 2 0,01 7 0,05 6

35% 4 1% 5 1,02 4 32 0

1986-1989

4

40726774

5 75% 3 0,02 6 0,07 5

53% 3 2% 4 1,06 3 29 0

1990-1995

6

511171353

3 67% 5 0,01 7 0,04 7 -46% 8 1% 5 1,19 2 37 0

1996-2000

5

1986000000

1 40% 9 0,1 2

2001-2004

4

1327750000

2 75% 3 0,02 8 -0,04 8 -126% 10 -2% 6 0,94 6 43 0

0,2

e

2

f

g

h

i

1382 % 1 9% 1 1,06 3 19 0

(Legend: a=average annual profit increase in €, b=occurrences of profit increase, c=ROA, d=ROE, e=profit increase % per employee, f=profit margin, g=asset turnover, h=total score, i=number of accounting innovations)

19

Several financial ratios have been used to assess performance in the investigated decades. More related ratios can be identified in table 4 of the appendix. Only the above stated relative performance indicators are used. The best performances are ranked ‘1’, second performance ‘2’, etcetera. When there is no information available in annual reports regarding certain ratios, the decade is awarded the most credits available. Periods are awarded ‘0’ in occasions where figures are used as a base year. When no base year is used the decade without any information regarding the selected measures will be awarded / punished with 1 credit more than the highest credit given in that decade. The decade with the lowest score is the best performing decade. This decade is compared to the decade with the most accounting innovations. The second column shows the annual profit increase on average generated in the decades in the first column. This is a good measure because it shows the money generating ability of the firm in general. Column ‘b’ counts the years in which profit increased compared to the previous year. It offsets the effect of one extraordinary year that could cause bias in the results of column ‘a’ and therefore rewards gradual growth. Columns ‘c’ and ‘d’ obviously provide the return on assets ratio and return on equity ratio15, which is a good measure for the efficient use of assets and equity. Column ‘e’ describes the average profit increase in percentages per worker. Column ‘f’ provides the average profit margin for every decade. Column provides the asset turnover ratios. Graph 1 of the appendix shows the dividend payout ratio over almost a century. The specific amounts are shown in table 3 of the appendix. The trend line reveals that the payout ratio in general is declining. When firms have low quality disclosures, but want to attract finances, they have to pay a high-risk premium. Therefore, high quality implies low dividend payouts. In order to make an assertion about innovativeness and the incentive to innovate, a distinction between -before and after 1950- is made. The development of the liabilities / equity ratio with the timing of the shift towards the closed model as identified in literature is reconciled. The ratio of equity to liabilities went from 89:11 in 1945-1946 to 31:69 in 1950. This information may inhabit bias due to the reintroduction of the consolidated annual reports in 1948. After the reintroduction of consolidated annual reports –which makes comparability better to 1950- leverage ratios were in 1948, 50:50 and in 1950 24:86. See also graphs 3 in the appendix.

3.5 Creative accounting & CEO succession From 1912 onwards, 11 periods are identified in which different CEO’s were in charge. Table 1a (p. 19) provides a more elaborate financial view on each CEO’s leadership. A distinction has to be made between non-routine (NR) and routine CEO turnover. Assessed in depth are non-routine successions by using four financial measures. In literature considerate evidence has been found that when approaching CEO turnover, income reducing accruals, research and development and depreciation

15

In this case ROA and ROE are calculated as follows: gross profit / assets or gross profit / equity

20

(directly measures the effect on income) are managed to affect the company result Detzler & Machuga, 2002, p 349-350). Using the current accruals instead of total accruals may increase the power of the test because firms are more likely to use discretionary current accruals to achieve year-specific earnings objectives (Detzler & Machuga, 2002, p 351). The big bath (cover up) accounting will be identified when R&D and depreciation expense’ grows (decline) towards a maximum (minimum) in year t and declines (increase) in year y+1. Additionally, the extraordinary gains / losses and accounting accrual should decrease (increase) in year t-1, hit a minimum (maximum) in year t and increase (decrease) in year t+1 again. Identified are growth rates for all non-routine CEO successions, concerning the four financial measures. In the cover up situation it is assumed is that the lead CEO in office can predict when he will be asked to resign. He therefore has more time to cook the books; this is the reason why a broader period is used. The amounts used for the calculations can be viewed in table 9 of the appendix 4. Table 4 provides all ratios concerning the above-specified measures. All measures are growth rates except for the accounting accruals. They express the growth rate of accounting accruals per euro of turnover. First, earnings management in general and routine and non-routine successions have to be identified. In the last forty years of the 20th century, seven CEO’s headed Philips:

F. Philips (1961-1970) Taking into account that surrounding the turnover of the previous CEO’s no suspicious earnings management occurred, the research starts in 1961. Considering that Mr. F. Philips became member of the board of commissionaires, this was a routine turnover. Therefore, no big bath or cover up incentive is present 1960/1961, although earnings have been managed.

H.A.C. Riemsdijk (1971-1976) Although in 1970 a decline in profits started to appear and continued till the end of 1971, a profit enhancing practice can be identified. According to Volmer and Faas the profit decline in 1971 should have been 38% whereas a 21% decline has been reported. The stated profit should have been 271 million, whereas the disclosed amount was 343 million (1997, p 100). Although the decline may come in handy for the new CEO, it cannot be concluded that a big bath had been taken. This occasion does not live up to the non-routine turnover requirement. considering there a routine succession. A definite conclusion is that earnings have been managed so that investors where provided with a more positive view of the company and its management.

N. Rodenburg (1977-1980) The difference in profit between 1979 and 1980 was remarkable. Appendix, Table 3 show the following profit numbers: fl. 619.000.000, - in 1979 and fl. 328.000.000, - in 1980 (amounts x 2, 2 because 1 euro=fl 2,2) Again Philips made accounting foundation changes in 1980; employee bonuses and profit shares were treated as being part of profit determination instead of profit distribution

21

(Hoogendoorn, 1992, p 378). At the end of his spell, Rodenburg was included in the board of commissioners.

W. Dekker (1981-1985) In 1980, employee bonuses and profit shares were treated as being part of profit determination instead of profit distribution (Hoogendoorn, 1992, p 378). As identified by Hoogendoorn 1981 was the year in which several policy changes occurred. The first change was that in compared to prior years a finance correction was used for value increases on assets. This was accompanied by the fact that regarding deferred taxes, the revaluation was corrected through the use of the profit and loss account and not via the revaluation reserve. Thirdly, currency differences -when recalculating equity of foleadership companies- are taken by the equity instead of by the usual revaluation reserve. Provisions for impairments of inventories and uncollectible accounts receivable are determined at the level of the estimated risk at the end of the book year, whereas provisions remained at a higher level in the passed. The provision for lifecycle duration risks in order to equalize lifecycle duration differences is omitted. Bought shares ‘N.V. Bezit’ subtracted from common stock instead of including at nominal value in the securities item (Hoogendoorn, 1990, p 378). The exact effect of the above mentioned cannot be determined, Hoogendoorn however, estimates that is has a positive impact on the 1981 result of somewhere between 703 and 1147 million Guilders. The disclosed result in 1981 is 357 million Guilders or 162 million in Euros; this means that Philips without a change in principles would have reported a loss between 346 million Guilders and 790 million Guilders. In 1980 a profit of 328 million Guilders was disclosed, the 1981 profit was intended to give the impression that Philips was getting back on track after the profit decrease between 1979 and 1980. In reality this was not the case (Hoogendoorn, 1990. p 379). Considering the routine character of the CEO succession, the above stated proves nothing more than earnings management.

C.J. van der Klugt (1986-1989) In 1985 the Philips annual report discloses a decline in profits, from fl 1.113.000.000,- in 1984 to fl 919.000.000 in 1985. The year after this event Philips’ profit climbed up again. At April 22, CEO Van Der Klugt was appointed. Several changes in principles have been identified, from which one of the below stated is profit increasing Costs regarding publishing rights were depreciated in 7 years instead of by using the ratio of expected revenues. WIR-premiums were directly booked in favor of profits when realization could reasonably be expected, instead of subtracting from the current value of assets. Deferred finance corrections were presented as being part of equity instead of being presented between equity and liabilities (Hoogendoorn, 1990, p 378). 1986´s profit was fl 1.015.000.000,- which is an increase of 10% in comparison to 1985. Although the proof for big bath accounting is minimal it is justified to conclude that earnings have been managed in a way that benefited CJ van der Klugt. If one would like to be persistent in finding big bath accounting than it can be argued that there was some big bath alike

22

approach. Since Van Der Klugt was appointed in April 1986, 1987 was his first full year. Subsequently he became a board member, which depresses the temptation to engage in big bath accounting.

J. Timmer (1990-1995) On April 10th 1990 Jan Timmer was appointed CEO of Philips. In 1989 the initial decline of 15% of profits was turned around into a solid performance improvement of 49%. In 1990 Philips made a turnaround which resulted in a big loss (fl. 4.240.000.000). It seemed that the shift in accounting principles was done in an attempt to gain some time to wait for better times. It became apparent that this would not happen and it was decided upon to make a clean sweep which resulted in the big loss that can be seen as taking a big bath. It cannot be proven that it was intentionally for the sake of the CEO’s reputation considering the loss in 1992. In 1992 a lot of accounting principles were altered this was to bring the companies disclosures more in line with international regulations (Volmer & Faas, 1997, p 100). Considering a non-routine succession occurred, this CEO turnover occasion will be investigated more in depth.

C. Boonstra (1996-2000) During Boonstra’s leadership a loss of € 268.000.000 was turned around into a profit of € 9.662.000.000 in the last year of his leadership. Considering his succession was not a routine one, it will be investigated in depth whether cover up or big bath accounting occurred. In 2001 G. Kleisterlee was his successor; this occasion was no routine succession either. Additionally, he also started with a loss as well. See also table 7 of the appendix.

23

4.

Hypopaper development

In order to find out if Philips was an accounting / disclosure innovator, it should disclose the predefined valuable innovations voluntary. This means that they have to appear earlier in Philips annual reports than legislation: H1. The accounting innovations appeared earlier in the annual reports than in legislation. A different way to assess Philips’ innovative spirit can be done by using Traas’ three ownership models (see introduction). The aim is to establish an assertion on the timing of Philips shift towards the open and closed model in comparison to the mainstream. Derived from literature is that the average mainstream model shift occurred in 1950 and 1970: H2a. Philips shifted definitively from proprietary ownership model towards closed model before 1950. H2b. Philips shifted definitively from closed ownership model towards open model before 1970. When a firm has a low quality of disclosures, it has to pay a risk premium to investors in order to attract finances. Assumed is that high quality implies low dividend payouts. The other side of the coin would imply that, in order to have high quality disclosures resources should be available. H3a. Accounting innovations occurred in high performance periods. H3b. Philips’ solvency ratio is negatively related to the accounting quality levels H3c. Philips’ dividend distribution is negatively related to the accounting quality levels The chronological description provides evidence to conclude that a need for finances was present and that this need occurred somewhere in the time period surrounding the swing to the closed model. An additional test will be done to add strength to hypotheses 3a and 3b and prove a need for finances. The following hypotheses are designed to do so: H3d. Equity < liability after the year 1950. H3e. Equity < liability before the year 1950. There are considerable disadvantages to performance related compensation plans surrounding CEO turnovers. Taking into account that in the last three CEO successions –which were all non-routine turnovers- profits decreased in the transition year, especially the likelihood of big bath accounting seems high. To test the level of creative accounting at Philips the following hypotheses are developed: H4a. Average discretionary expenses (income) in year t-1 and t-2 < (>) average discretionary expenses in years t + 1 and t + 2 (cover up). H4b. Discretionary expense (income) in year t-1 < (>) discretionary expense (income) in year t (big bath).

24

5. Results and discussion (1) Innovation hypotheses Out of the 12 innovation opportunities, Philips was earlier with applying those innovations in 11 occasions. Next to this, the 1 occasion in which it was not an innovator was in disclosing deferred tax, as stated in table 10 of the appendix. As could be identified in literature, this item could be inferred, but was not required by law. This makes our conclusion even more solid. Hypopaper 1: ‘the predefined innovations were disclosed voluntary by Philips’ is supported. Another conclusion is that P.F.S. Otten was responsible for the innovative edge Philips had during the 1950’s with a 50% score as stated in table 6 of the appendix.

(2) Ownership model hypotheses Considering the voluntary improvements in disclosures and accounting practices in the late 30’s, 40’s and the reintroductions of innovations directly after the Second World War and shortly before Philips became listed on the New York stock exchange it can be concluded that the practices lived up to the demands of the closed model and started before the year 1950 at Philips. Especially when aspects that were found when investigating for other hypotheses are incorporated. Particularly the findings of 4.3; that Philips was financed with more liabilities than equity before 1950 adds strength to the evidence. Therefore, H2a. ‘Philips shifted definitively from proprietary ownership model towards closed model before 1950’ is supported. Philips was working towards or at least cooperating to arrive at the open model situation already before others were willing to do so. This is a conclusion to be drawn from before discussed information. Especially when considering that the NIvRA’s meeting -in which Philips’ auditor played a big roledate was April 1967 (Zeff et al, 1992, p 188), the first annual report was published in 1968, the auditors profession became more on behalf of the public interest and the membership of a Philips functionary in the CAJ. The involvement of Philips functionaries in subsequent periods adds strength to the hypopaper. Therefore H2b. ‘Philips shifted definitively from closed ownership model towards open model before 1970’ is supported.

Recapitulating; both when assessing innovations individually and using the ownership approach the conclusion can be drawn that Philips was earlier with implementation than the mainstream or when legislation required it. Moreover, Philips internal auditor was one of the founders of the CAJ, the Tripartite Study Group’s predecessor. Subsequently, several Philips functionaries were active in the Tripartite Study Group after it was set up. This makes Philips and its management not only an innovator but also a public sparring partner concerning accounting and disclosure policies. Furthermore this implies that not only P.F.S. Otten was engaged with accounting improvements but both H.A.C. Riemsdijk and the late Frits Philips as well.

25

(6) Disclosure & agency theory hypotheses In table 1b is stated that the decade with the best performance is the same decade as the one in which the most accounting innovations occurred. The 1950’s ranked first with 5 out of 8 occasions. This decade leaves the rest behind with 1912-1919 and 1990-1999 as runners-up with only 2 out of 8. This supports hypopaper 3a and implies that during times of high performance resources were made available to make the innovations possible. Whether quality of disclosure is negatively related to dividend payout is not very strong supported. When taking a ‘CEO approach’ an interesting phenomenon is identified, the most innovative CEO paid out the lowest dividends of the period 1912-1990 (see Table 7). Nevertheless, Table 8 of the appendix splits the periods up in decades, this does not allow for concluding that during the most innovative period, the dividends were the lowest. When limiting the time period to 1912-1979 (because 1979 was the final year of the decade in which the identified accounting innovations occurred for the last time), the negative relationship between innovations and solvency ratio (equity/debt) is identified. The trend lines and CEO-classification provide some support for hypopaper 3b but it is weak. When splitting up Philips history in decades, graph 3 of the appendix shows that the liability/ and equity-line cross during the 1950´s or liabilities> owners’ equity.

A more detailed view shows that

already before the mainstream shift towards the closed model Philips became more financed with capital from outside the company than from inside the company. To be exact the breakpoint was in 1948 during the leadership of Otten, as can be seen in Table 3 of the appendix. This means that H3d. Equity < liability after the year 1950 is not supported but H3e. Equity < liability before the year 1950 is. However, the need for finances was an incentive to shift towards the closed model which made internal financing by revaluation possible. Therefore, the accounting innovations (1950’s) and the need for (internal) financing (1950’s) (also identified by Brink, 1992, p 259) and the closed model (1950) are interrelated. Conclusion: H3c is supported. Recapitulating; resources were made available to improve accounting and disclosure quality during the 1950’s. A few years after the second World War the shift towards the closed model could be identified. In this period, Philips paid out the lowest level of dividends of the period 1912-1990. This supports the internal financing hypopaper 3c. However, weak evidence was found for the assertion that the solvency ratio is negatively related to accounting quality in general. In 1948 Philips was financed with more ‘outside capital’ than ‘inside (shareholders) capital’, because the need on capital for expansion was immense.

(4) Creative accounting hypotheses The calculations surrounding 2001, point out that surrounding Kleisterlee’s first full year a big bath was taken. From the three non-routine CEO successions this was the only one in which four out of four items supported the big bath approach. No evidence for cover up accounting surrounding Cor Boonstra succession was found. The calculations regarding Cor Boonstra’s succession of Jan Timmer in 1996, showed mixed outcomes. Evidence was found for cover up accounting. This means that Jan Timmer

26

sought for deferral of his resign. Additionally, Cor Boonstra undertook measures to boost his own performance by using accounting accruals and extraordinary gains and losses to present a more positive picture of his own performance. The succession of Van der Klugt by Timmer brought mixed results regarding big bath accounting. Accounting accruals and R&D expenses are suspicious which is supporting evidence for big bath hypopaper. Especially the significant decrease in R&D expense –as can be seen in table 9 of the appendix- in 1991 adds strength to this hypopaper. No evidence for cover up accounting was identified. This can be seen as obvious, taking into account that Van der Klugt was made member of the board of commissionaires. This adds power to the proposition of selecting only non-routine succession for this part of the paper.

6 Conclusion, implications and limitations During Philips’ high performance decade P.F.S. Otten led the firm. He made resources available to improve accounting and disclosure quality in order to attract finances to rebuild the company after the Second World War. A few . It is possible to identify the shift towards the closed model in the years after the second World War. A development that underlines the need for internal financing once more. Without a need for finances, the incentive to shift would have been non existent. Even before the 1950s Philips was financed with more ‘outside capital’ than ‘inside capital’. In this period, Philips paid out the lowest level of dividends of the period 1912-1990. One could argue low pay-out was not wise when trying to attract investors. On the other hand, the resources were needed to invest in the company; assets and production grew rapidly. Additionally, investors were provided with high quality disclosures and get more trust in Philips’ corporate governance and therefore willing to accept lower profit payouts. Little evidence was found for the assertion that the solvency ratio is negatively related to accounting quality in general. The title ‘accounting innovator’ was well deserved. Both when assessing innovations individually and using the ownership approach a strong conclusion can be drawn. Philips implemented innovative accounting procedures earlier than most companies and before legislation. Moreover, Philips’ internal auditor was one of the founders of the CAJ, the Tripartite Study Group’s predecessor. Subsequently, several Philips functionaries were active in the Tripartite Study Group after it was set up. This makes Philips and its management not only an innovator but also a stimulating partner concerning accounting and disclosure policies. Furthermore, this implies that not only was P.F.S. Otten engaged with accounting improvements but both H.A.C. Riemsdijk and Frits Philips as well. Throughout Philips history, it is possible to identify several earnings management events. This research has shown that the accounting practices surrounding Gerard Kleisterlee’s succession of Cor Boonstra were suspicious and point in the direction of big bath accounting. Evidence was found that suggested that big bath was created between the transition year of Boonstra and Timmer. Only in the starting year of Kleisterlee, did all measures support the big bath hypopaper.

27

During Boonstra’s 5 year spell at Philips, he managed to generate the highest average annual profit increase (big bath) by increasing Philips’ profit margin and profits per employee. The highest average profit was accomplished and amounted to € 2,835,104,396. Boonstra was the best performing CEO when the years he was in charge are assessed and compared with other CEO’s. When assessing the different decades of Philips’ history, decade wise, the 1950’s were the most successful years. In that period P.F.S Otten was in charge. He directed the firm towards the most gradual annual profit growth and asset turnover in Philips history. Jan Timmer was second in terms of runner-up in the asset turnover was. As mentioned before, leverage increased right after the Second World War, but the highest percentage of leverage was observed during Jan Timmer’s spell with only 24% of equity. Not only was the highest return on assets generated by A.F. and G.L.F. Philips, 22%, also the ROE was the highest during their CEO-ship. Throughout this period, the highest amounts of dividends were paid: 67%. During Riemsdijk’s leadership between 1971 and 1976 turnover increased by 173% on average. For the duration of his time in charge, Philips’ workforce reached its peak with 390,167 employees on average (see Table 3 and graph 4 in the appendix). It shows the need for growth after the Second World War and the turnaround in the 1970s that was characterized by downsizing, automating and, most of all a shift towards efficiency instead of sheer growth. As can be seen in graph 3 (and Table 3) of the appendix there was a gradual increase of assets over time. The highest level of assets was reached in the 2001-2004 period which can be attributed to Gerard Kleisterlee. Since the focus of this research starts in 1912, the years that A.F. and G.L.F. Philips were in charge of the company are limited to 10. From Table 7 of the appendix a clear trend towards shorter CEOleadership (with manipulation risks) can be viewed. Otten was a high performing CEO and stayed in charge of Philips for a long time while Boonstra’s period was successful as well, based on the prespecified measures, his working relationship with Philips was relatively short. Further research is necessary with regard to hypopaper three, little evidence was found for the assertion that dividend payout is negatively related to accounting quality in general. An investigation focused on assessing this is required. Another suggestion would be to assess Philips’ performance on other financial measures. A valuable inclusion could be cash flow information that have recently been judged as more valuable than profits. The creative accounting hypopaper might be more strongly supported by including more financial measures. Over the whole period the average pay out amounts to 36%. The accounting innovations can be viewed from a different window too. In booming years Philips accounting/reporting policy tried to reduce profit by creating hidden reserves in the first half of the period with low disclosures and open reserves (and provisions) in the second half with full disclosures. In crisis periods Philips tried to use these hidden/open reserves to prevent losses in order to survive.

28

References

AAA Financial Accounting Standards Committee (2002). Recommendations on disclosure of nonfinancial performance measures. Accounting Horizons, vol. 16 No 4. December 2002, 353-362

Barth, M.E & Murphy, C.M. (1994). Required financial statements disclosures: purposes, subject, number, and trends. Accounting Horizons, December 1994, 1-22

Beaver, W.H (1978). Current trends in corporate disclosure, future disclosure requirements may give greater recognition to the professional community. The journal of accountancy, January 1978, 44-52

Belkaoui, A. & Cousineau, A. (1977). Accounting information, nonaccounting information, and common stock perception. Accounting Information, July 1977, vol. 50, no. 3, 334-342

Botosan, C.A. (2002). A re-examination of disclosure level and the expected cost of equity capital. Journal of accounting research, vol. 40, no.1, March 2002, 21-40

Brink, H.L. (1992). A history of Philips’ accounting policies on the basis of its annual reports. European Accounting Review, 1992, 255-275

Cahan, S.F, Rahman, A, Perera, H, (2005). Global diversification and corporate disclosure. Journal of international accounting research, vol. 4, no. 1, 2005, 73-93

Camfferman, C. (1996). Voluntary annual report disclosure by listed Dutch companies 1945-1983. Ridderkerk: Offsetdrukkerij Ridderprint.

Chatfield, M. & Vangermeersch, R. (1996). The history of accounting, an international encyclopedia. New York: Garland

Choi, D.S. (1973). Financial disclosure and entry to the Europeans capital market. Journal of accounting research, autumn, 1973, 150-175

29

Cooke, T.E. (1991). An assessment of voluntary disclosure in the annual reports of Japanese corporations. The international journal of accounting, 26, 174-189 Dechow, P.M, & Sloan, R.G. & Sweeney, A.P (1996). Causes and consequences of earnings manipulation: an analysis of firms subject to enforcement actions by the SEC. Contemporary accounting research, vol.13, no.1 spring 1996, 1-36 Enthoven, A.J.H., , Current Value Accounting, its concepts and Practices at Philips’ Industries, The Netherlands, International Research Study # 3, 1982 Greenstein, M. M. & Sami, H. (1994). The impact of the SEC’s segment disclosure requirements on bid-ask spreads. The accounting review, vol 69, no.1, January 1994, 179-199 Hoogendoorn, M.N (1990). Change in Companies’ accounting Principles. Groningen: WoltersNoordhoff bv, 1990 Johnson, L.T. (1992). Research on disclosure. Accounting Horizons, March 1992, 101-103 Kirschenheiter, M. & Melumad, N.D. (2002). Can ‘big bath’ and earnings smoothing co-exist as equilibrium financial reporting strategies? Journal of accounting research, vol 40. no. 3 june 2002 Lee, T.A. (1983). Company financial reporting. (Second edition). Berkshire: Van Nostrand Reinhold Co. Ltd Peek, E. (2004). The use of discretionary provisions in earnings management: evidence from The Netherlands. Journal of international accounting research, vol 3, no 2 2004 Riahi-Belkaoui, A. (2004). Accounting theory. (fifth edition). London: Thomson, 2004 Schattke, R.W., Vergoossen, R.G.A., Barriers to Interpretation-A Case Study of Philips Electronics NV, Accounting and Business Research, vol. 27, No. 1, pp.72-84, 1996 Schroeder. R.G. & Clark, M. (1995). Accounting theory: text and readings. (fifth edition). New York: John Wiley & Sons, Inc Stanga, K.G. (1976). Disclosure in published annual reports. Financial management, winter 1976, 4250 Sutton, T. (2000). Corporate financial accounting and reporting. Edinburgh: Pearson Education Limited. Vangermeersch, R., An historical Analysis of Financial Reporting Practices of Financial Reporting Industries of the Netherlands for Tangible Fixed Assets and Inventory (1915-1981), The Academy of Accounting Historians, Working Paper III (57), 1983 Volmer, F.G. & Faas, F.A.M.J. (1997). External Financial Reporting: A New Approach. InterFaas Research and Advice, Boekel, 1997 Zeff, S.A. & Well Van der, F. & Camfferman K. (1992) Company financial reporting, A historical and comparative study of the Dutch regulatory process. Amsterdam: North-Holland

30

Appendices ( Royal Philips 1912-2000)

Table 3 Important Philips data Table 4 Cover up accounting Table 5 Big bath accounting Table 6 Timing of accounting innovations and legislations Table 7 Years CEO in office, payout and accounting innovations Table 8

Decades, payout and accounting innovations

Table 9 Accruals and non-routine CEO succession

Graph 1

size of financial report

Graph 2

turnover development

Graph 3

assets, stockholders’ equity and liabilities

Graph 4

number of employees

Graph 5

number of balance sheet items

Graph 6

number of profit and loss items

31

32