Sharing the pain? An examination of CEO and executive compensation of publicly traded newspaper companies

Abstract

This study examines the compensation of newspaper company chief executive officers (CEOs) and other top executives, comparing compensation with key measures of the companies’ financial performance and employment levels. Fixed-effect regressions found only a small relationship between CEO pay and companies’ market value for 2000 to 2013. There was no relationship between pay and return-on-assets or return-on-equity. Unobserved characteristics of individual companies are associated with CEO pay. The implications for the financial health of newspaper companies are discussed.

Keywords

newspaper companies executive compensation corporate performance

Large, publicly traded newspaper companies have experienced significant declines in revenue and employment (Barthel, 2018). However, the performance of chief executive officers (CEOs) and other top executives at these companies has not been widely studied. These executives are ultimately responsible for their companies’ finances that enable the provision of news coverage to communities where their newspapers operate. In many communities throughout the United States, the declining local newspaper is still the major source of local news. Digital-only news publications have not been able to replace the jobs, news coverage, and advertising that are being lost (Barthel, 2015; Fico et al., 2013; Holcomb, 2015; Jurkowitz, 2014).

This study examines how the financial declines at publicly traded newspaper companies influenced executive compensation at those companies. Economists and business theorists have long argued that the best way to ensure good financial performance is to tie executive compensation to the profitability of the company the executive manages (Eisenhardt, 1989; Hou, Priem, & Goranova, 2014; Murphy, 1999). This study examines the compensation of CEOs and other top executives at publicly owned newspaper companies during the last period when the companies were financially healthy and during the period when those companies experienced their steepest financial declines. The study compares executive compensation with key measures of each company’s financial performance, and with employment at the companies the executives managed. The study finds that executive compensation at these companies had either a weak or non-existent relationship to measures of financial health and employment.

Overview of CEO Compensation

The amount of CEO compensation paid by U.S. companies is a contentious topic among the public, politicians, and regulators. The president of the United States has even criticized what he considered inordinately high compensation paid to some CEOs (Bhagat & Romano, 2009). Financial reform legislation passed in 2010 requires that companies’ executive compensation practices be approved or disapproved by shareholders in a non-binding vote at least once every 3 years (Barry & Kairis, 2011; Dodd-Frank Wall Street Reform and Consumer Protection Act, 2010).

The board of directors sets CEO compensation for publicly traded companies. A subcommittee of the board—the compensation committee—is charged with reviewing CEO performance and recommending to the full board the CEO’s compensation package. The compensation committee is composed entirely of independent members of the board of directors. U.S. Securities and Exchange Commission (SEC) regulations and stock market rules define independence (Listing Standards for Compensation Committees, 2012; Leff et al., 2015). Committee members should not receive compensation from the company that is large enough to “impair the Committee member’s ability to make independent judgments about the Company’s executive compensation” (Leff et al., 2015) or have any relationships that place them under the “direct or indirect control of the Company or its senior management” (p. 79).

In a company’s DEF-14A filing with the SEC, the compensation committee must detail for shareholders the method for reviewing CEO performance and determining the amount and nature of the CEO’s compensation. For most publicly traded companies, the compensation committee is advised by a consulting firm that specializes in CEO reviews (Downs, 2012). Because of changes in the tax law, CEO performance reviews are tied to specific measurable financial goals such as return on stockholders’ equity, return on investment, net income levels, and increasing the company’s market value, which is based on stock price. Each company is free to establish its own financial measures for evaluating CEO performance, and these measures can vary widely from company to company and from industry to industry. The compensation committee also examines the compensation of CEOs in the company’s peer group. These are companies in a similar business that may or may not be direct competitors of the company. The peer group of companies used for comparison is identified in the DEF-14A filing.

This study examines levels of executive compensation in publicly traded newspaper companies, paying particular attention to compensation levels when the newspaper industry went into financial free fall in 2005. The conceptual context underpinning this study is the research in management and finance that focuses on executive compensation and how compensation amounts are determined. Two overarching issues are our focus: (a) what happened with executive compensation at publicly traded newspaper companies during the industry’s financial collapse? and (b) did executive compensation follow what agency theory predicts, that is, compensation being pegged to a company’s financial performance?

CEO Compensation

Compensation paid to CEOs of U.S. companies has been increasing much faster than the compensation of the average worker. In 1970, CEO compensation averaged US$700,000, or 25 times the compensation of the average worker (Downs, 2012). By 2000, the multiple for CEOs had increased to 90 times the average worker’s compensation. The average compensation for CEOs reached US$11.7 million in 2013, while the average worker earned US$35,239. The average CEO’s compensation is now 331 times that of the average worker (American Federation of Labor and Congress of Industrial Organizations [AFL-CIO], 2014).

Defenders of high levels of CEO compensation point to the following: (a) the need to recruit and retain skilled managers in a competitive market; (b) tying compensation to financial goals to incentivize executives to focus on meeting or exceeding these goals; (c) compensation with stock and stock options incentivizes CEOs to work in the best interest of their shareholders by increasing shareholder value; (d) longevity of CEOs results in higher compensation; and (e) most CEOs come up through a company’s ranks, which results in higher levels of compensation (Downs, 2012).

However, public concern that executive compensation reflects corporate greed has resulted in legislation to limit executive compensation. In 1993, Congress changed the tax law to cap CEO compensation. Previously, a company could take a tax deduction for the full amount of a CEO’s compensation. The change capped CEO compensation at US$1 million. Any compensation above this amount had to be tied to specific financial performance goals set by the board of directors, and these goals had to be in the company’s annual proxy filing. By doing this, Congress believed it was giving shareholders more say in CEO compensation. Board members are elected by shareholders, and Congress expected shareholders would elect board members who were sympathetic to shareholder ideas about fair and adequate compensation. However, much of the stock in public companies is managed by mutual funds, banks, and other large institutional investors. Individual shareholders tend not to vote their shares, allowing these institutions to vote them instead (Cranberg, Bezanson, & Soloski, 2001). There is no reason to assume that those voting the shares have the same goals as those who actually own the shares (Downs, 2012).

Companies can also circumvent the US$1 million cap. A few companies simply forgo the tax deduction and pay the extra income tax. But most companies changed their executive compensation packages by shifting CEO compensation away from salary and bonuses and toward stock, stock options, and deferred compensation, which are not considered compensation for income tax purposes. These other forms of compensation are subject to personal income tax, but most U.S. companies pay this tax themselves.

The next major reform was enacted by the SEC, which regulates public companies, in response to very large compensation packages of some highly visible CEOs and the collapse of Enron and WorldCom because of complicity among top executives. To increase transparency, the SEC in 2006 required all publicly traded companies to report compensation of the company’s five highest paid executives using seven specific categories: (a) salary, (b) bonus, (c) stock grants, (d) stock options, (e) non-equity incentive plan, (f) pension and deferred compensation, and (g) other compensation.

The SEC also mandated that the board of directors make more transparent how executive compensation was determined. However, a company’s board of directors retains discretion to determine the amount and nature of CEO compensation. Thus, in its DEF-14A filing, a company spells out the financial measures used to evaluate CEO performance, but the company does not have to disclose how specific measures contributed to setting the amount of compensation.

As noted earlier, companies that trade their stock on the New York Stock Exchange and NASDAQ must have independent members of their board of directors on their compensation committees (Listing Standards for Compensation Committees, 2012; Leff et al., 2015). Independent board members are supposed to be less subject to pressure and influence from the CEO. The stock exchanges also specify the criteria a board member must meet to be considered independent. As long as a former executive or employee has not been employed by the company for the past 3 years, she or he is considered an independent director. Two newspaper companies, McClatchy and Media General, have directors who meet the requirements for independence even though McClatchy’s board chairman is a member of the company’s founding family and, in the case of Media General, an independent director was its long-time CEO.

All newspaper companies tout the importance of journalism, but journalism does not appear to play much, if any, role determining executive compensation. Cranberg et al. (2001) found that only one publicly traded newspaper company assessed the quality of journalism when reviewing the performance of top management. They concluded that, “By and large, so little attention is paid to a company’s brand of journalism that it is possible to study a committee report of how top managers are rewarded and not to realize that the enterprise is engaged in journalism” (p. 49). Compensation committee members are independent directors, so it is not surprising that very few have any background in journalism. On committees populated with bankers, lawyers, and business executives, financial data drive compensation, not winning journalistic prizes.

The first two questions addressed in this article are as follows:

RQ1: How much compensation are newspaper companies’ CEOs and other top executives receiving and what is the nature of that compensation? That is, how much of the total compensation package comes from salaries, bonuses, stock, stock options, and deferred compensation?

RQ2: As publicly traded newspaper companies saw both their book and market values drop precipitously in the mid-2000s, how did this affect executive compensation? Did CEOs and other top executives share the financial pain of their investors?

Setting the Stage: Newspaper Industry’s Financial Collapse

The first decade of the 21st century marked the very best and the very worst of times for the newspaper business. Within 5 years, the industry hit both its all-time financial highs and all-time financial lows. The stunning financial collapse that occurred in the mid-2000s and continues to this day was the result of complex factors. Press critics, journalists, scholars, and newspaper companies agree that management did not position the companies to deal with the challenges posed by the Internet. Managers also failed to position their companies to take advantage of the business opportunities provided by the Internet (Cho, Smith, & Zenter, 2016; Collis, Olson, & Furey, 2010; Cranberg et al., 2001; Karimi & Walter, 2016; Soloski, 2013).

In the mid-2000s, the newspaper industry had to face a new business reality: sustained and permanent losses in its major source of revenue. Previously, a down year in advertising revenue was almost always followed by an up year. Between 1985 and 2005, print advertising revenues increased at an average annual rate of US$1.1 billion (Edmonds, Guskin, Rosenstiel, & Mitchell, 2012). The apogee was 2005 when advertising revenue topped out at US$47.4 billion. Since then, advertising revenue has been in a free fall. In just 6 years, 2005 to 2011, newspaper advertising revenue dropped 56.4% or US$26.7 billion.

Table 1 reports the market value and the stock prices for the six major public newspaper companies that existed in 2012. (Market value is reported by the companies in their 10-K filings with the SEC. Companies must give a date for setting the market value. Share price is the closing stock price on that day or on the last business day before that date if the date falls on a weekend or holiday.) Between 2000 and 2004, the market value of the six newspaper companies increased 38.6% from US$27.4 billion to US$38 billion. From both the revenue side and market side, 2004 was the newspaper industry’s banner year. But the bottom fell out quickly. In 2005, the market value of the companies dropped US$7.7 billion or 20.2%. This was not a normal market correction; the market value of these companies continued to plummet (Soloski, 2013).

Table 1

Share Price and Market Value

	2012	2011	2010	2009	2008	2007	2006	2005	2004	2003	2002	2001	2000
Share price (US$)
Post	363.55	$422.21	$408.25	$400.30	$584.46	$776.09	$780.01	$849.25	$896.10	$733.00	$545.00	$571.78	$621.50
Lee	1.15	$2.68	$3.16	$0.32	$10.76	$29.86	$33.29	$38.42	$47.68	$43.02	$34.28	$35.25	$28.31
Gannett	13.5	$13.53	$14.88	$3.73	$22.02	$54.95	$54.17	$72.06	$84.95	$76.59	$75.90	$61.98	$61.17
NYT	6.55	$7.98	$9.40	$5.42	$15.65	$25.40	$23.82	$31.08	$44.95	$47.17	$51.50	$44.93	$43.55
McClatchy	2.17	$2.63	$4.28	$0.77	$7.19	$25.31	$41.60	$65.39	$69.60	$57.18	$64.25	$57.75	$38.75
Media General	4.01	$4.03	$10.80	$1.88	$13.17	$33.27	$40.41	$65.00	$63.92	$56.89	$60.00	$57.74	$50.39
% change in stock price
Post	−13.9	3.4	2.0	−31.5	−24.7	−0.5	−8.2	−5.2	22.3	34.5	−4.7	−8.0
Lee	−57.1	−15.2	887.5	−97.0	−64.0	−10.3	−13.4	−19.4	10.8	25.5	−2.8	24.5
Gannett	−0.2	−9.1	298.9	−83.1	−59.9	1.4	−24.8	−15.2	10.9	0.9	22.5	1.3
NYT	−17.9	−15.1	73.4	−65.4	−38.4	6.6	−23.4	−30.9	−4.7	−8.4	14.6	3.2
McClatchy	−17.5	−38.6	455.8	−89.3	−71.6	−39.2	−36.4	−6.0	21.7	−11.0	11.3	49.0
Media General	−0.5	−62.7	474.5	−85.7	−60.4	−17.7	−37.8	1.7	12.4	−5.2	3.9	14.6
Market value
Post	2,300,000	3,200,000	3,700,000	2,800,000	2,600,000	4,500,000	4,300,000	4,700,000	4,927,000	3,984,000	2,881,000	3,030,000	2,954,000
Lee	57,507	114,069	141,794	12,576	445,331	1,285,192	1,450,973	1,617,843	1,995,243	1,881,719	1,493,437	1,515,200	1,197,960
Gannett	3,123,094	3,252,839	3,544,790	874,868	5,020,320	12,810,729	12,796,617	17,609,506	22,717,076	20,590,444	20,121,013	20,110,302	15,724,453
NYT	961,000	1,100,000	1,300,000	734,000	2,100,000	3,400,000	3,200,000	3,600,000	5,420,000	5,470,000	6,380,000	5,500,000	5,800,000
McClatchy	135,300	171,600	257,100	281,000	426,400	1,500,000	2,365,373	1,400,326	1,488,467	1,191,641	1,308,630	1,165,794	762,889
Media General	89,755	89,259	237,959	40,526	282,967	693,923	901,500	1,443,600	1,405,200	1,232,500	1,390,000	1,165,000	996,000
Tribune						3,480,700	6,998,000	8,270,000	10,739,000	11,249,000	9,469,000	9,310,000	8,508,000
Journal Register						175,367	335,636	722,853	803,900	714,009	567,167	684,511	271,160
Total	6,666,656	7,927,767	9,181,643	4,742,970	10,875,018	27,845,911	32,348,099	39,364,128	49,495,886	46,313,313	43,610,247	42,480,807	36,214,462
Total six solvent companies	6,666,656	7,927,767	9,181,643	4,742,970	10,875,018	24,189,844	25,014,463	30,371,275	37,952,986	34,350,304	33,574,080	32,486,296	27,435,302
% change	−15.9	−13.7	93.6	−56.4	−60.9	−13.9	−17.8	−20.5	6.9	6.2	2.7	17.3
% change in market value
Post	−28.1	−13.5	32.1	7.7	−42.2	4.7	−8.5	−4.6	23.7	38.3	−4.9	2.6
Lee	−49.6	−19.6	1,027.5	−97.2	−65.3	−11.4	−10.3	−18.9	6.0	26.0	−1.4	26.5
Gannett	−4.0	−8.2	305.2	−82.6	−60.8	0.1	−27.3	−22.5	10.3	2.3	0.1	27.9
NYT	−12.6	−15.4	77.1	−65.0	−38.2	6.3	−11.1	−33.6	−0.9	−14.3	16.0	−5.2
McClatchy	−21.2	−33.3	−8.5	−34.1	−71.6	−36.6	68.9	−5.9	24.9	−8.9	12.3	52.8
Media General	0.6	−62.5	487.2	−85.7	−59.2	−23.0	−37.6	2.7	14.0	−11.3	19.3	17.0
Tribune						−50.3	−15.4	−23.0	−4.5	18.8	1.7	9.4
Journal Register						−47.8	−53.6	−10.1	12.6	25.9	−17.1	152.4

Note. NYT = New York Times.

By any financial measure for assessing a company’s return on stockholders’ equity, newspaper companies failed their investors during the first decade of the 21st century (Soloski, 2013). By 2012, investors in newspaper companies would see almost all of their investments wiped out.

By 2009, the total market value of the six viable newspaper companies was US$4.7 billion, compared to US$38 billion in 2004. In just 5 years, these companies lost an incredible US$33.2 billion or 87.4% of their market value. The drop in market value of individual companies was enormous. Between 2004 and 2012, Gannett’s market value dropped US$19.6 billion or 86.3%, the New York Times Co. lost US$4.5 billion or 83% of market value, and the Washington Post Co. lost US$2.6 billion or 53% of market value. Between 2004 and 2012, Media General lost 93.6% or US$1.32 billion of its market value. Between 2004 and 2012, Lee Enterprises went from a company worth US$2 billion to one worth US$57.5 million, a drop of over 97% in market value. Tribune Co. filed for bankruptcy in 2007 and the Journal Register Co. did in 2008, together wiping out US$12 billion in shareholders’ money.

Examining stock prices in Table 1 drives home the impact on investors of the industry’s collapse. Beginning with the year of each company’s highest stock price, in 2004, Gannett’s stock was trading at US$84.95 per share; by 2009, the price had plummeted to US$3.73, a drop of 95.6%. Lee Enterprises saw its stock drop from a high of US$47.68 to 32 cents in 2009, a drop of 99.3%. The drop caused the New York Stock Exchange to warn Lee Enterprises that its stock faced delisting. (Delisting stock makes trading more difficult. Some large institutional investors will not own stock that is not traded on a major exchange.) The New York Stock Exchange warned McClatchy twice that its stock faced delisting after its stock price dropped to 77 cents in 2009, down from a high of US$69.60 in 2004. Media General’s stock went from US$65 a share in 2005 to US$1.88 in 2009, a drop of 97%. The price of New York Times Co. shares went from US$51.50 in 2002 to a low of US$5.42 in 2009, a drop of 89.5%. The Washington Post Co. was the only newspaper company that did not experience such a huge drop in stock price. Even so, its stock dropped from a high of US$896.10 in 2004 to US$363.55 in 2012, a drop of 59.2% (Soloski, 2013).

In addition to plummeting market valued, these companies experienced large drops in their book value (the value of assets minus depreciation, as contrasted with market value, which is the number of outstanding shares times the price of those shares), large increases in their debt loads, and significant increases in goodwill as a percentage of their total assets, which the companies were forced to write off, causing large drops in their book value (Soloski, 2013). Three companies had to eventually file for bankruptcy and others were precariously close to bankruptcy. During some years, a number of the companies were underwater, with liabilities exceeding assets (Soloski, 2013).

As shareholders were suffering, so, too, were the companies’ employees. In 2006, the newspaper companies employed 104,511 people. CEOs are responsible for employees throughout their companies, not just those who work in newsrooms. Public companies in this study reported total employment but did not provide breakdowns by job category. (Employment numbers are from the companies’ 10-K filings. Media General did not publish employment data each year and is not included in the employment data reported in this study.) By 2012, that number had dropped 41.5% to 61,209 employees. In 2004, just before the advertising market collapsed, Gannett employed 53,000 people, a number that dropped to 31,600 by 2012. Within 8 years, Gannett reduced its workforce by 40.4%. The other newspaper companies also shed a large number of employees. The Washington Post employed 21,500 people in 2009. By 2012, the number had shrunk to 14,000, a drop of 34.9%. Lee Enterprises cut its workforce by half between 2005 when it had 10,000 employees and 2012 when it employed 5,000. The New York Times employed 12,300 people in 2004, but by 2012, it employed 3,500 people, a drop of 71.5%. In 2007, McClatchy employed 15,748 people; by 2012, the number of employees was down 55% to 8,668 employees.

There are two additional questions addressed by this article:

RQ3:One consequence of the financial collapse of newspaper revenues in the mid-2000s was that a large number of employees were laid off and some companies ordered mandatory furloughs for employees and froze salaries. Did newspaper company CEOs share the financial pain of their employees?

RQ4: As newspapers’ market value stabilized, albeit at near record lows, was this reflected in executive compensation?

Newspaper Company CEO Compensation

In 2012, newspaper CEOs earned an average of US$4.4 million in total compensation according to Table 2. Gannett’s CEO earned the most, US$8.5 million, and the Washington Post’s CEO the least, US$880,000. Donald Graham, the Post’s CEO, had frozen his annual salary at US$400,000. He also received the least of any CEO in stock, stock options, and deferred compensation. If the Post is removed, the average CEO compensation for newspaper companies rises to US$5.0 million.

Table 2

CEO Compensation

	% salary and bonus	% stock and stock option	Salary and bonus (US$)	Total compensation (US$)
Gannett
2012	29.4	34.7	$2,482,692	$8,453,598
2011	5.3	15.1	$650,000	$12,255,114
2010	29.0	53.6	$2,730,769	$9,405,049
2009	50.9	39.2	$2,392,308	$4,698,692
2008	65.1	30.3	$2,041,667	$3,135,469
2007	39.1	58.6	$2,950,000	$7,546,710
2006	36.2	45.6	$2,950,000	$8,150,444
2005	47.3	51.4	$3,950,000	$8,345,621
2004	34.7	63.8	$4,050,000	$11,666,214
2003	29.9	68.6	$3,850,000	$12,879,673
2002	30.6	68.3	$3,850,000	$12,575,912
2001	28.2	70.8	$3,450,000	$12,221,827
2000	29.8	69.3	$3,058,000	$10,271,669
McClatchy
2012	10.9	18.4	$489,462	$4,491,610
2011	22.4	33.7	$975,000	$4,353,634
2010	25.7	0.0	$953,700	$3,705,528
2009	26.0	41.0	$976,250	$3,753,229
2008	39.7	42.2	$1,100,000	$2,767,306
2007	23.8	43.4	$1,100,000	$4,618,167
2006	36.6	32.8	$2,050,000	$5,601,811
2005	55.3	24.3	$1,950,000	$3,528,752
2004	81.4	17.7	$2,050,000	$2,519,965
2003	55.0	43.2	$1,725,000	$3,133,686
2002	51.9	46.2	$1,600,000	$3,083,156
2001	41.3	56.2	$1,125,000	$2,721,926
2000	55.9	42.2	$1,200,025	$2,148,256
Media General
2012	29.0	18.5	$938,700	$3,236,544
2011	58.7	8.1	$844,544	$1,438,540
2010	47.2	38.5	$938,700	$1,986,713
2009	64.6	2.8	$871,635	$1,350,249
2008	33.4	36.6	$925,000	$2,767,867
2007	29.2	28.1	$585,000	$2,002,133
2006	23.1	24.9	$730,000	$3,165,222
2005	24.4	62.4	$1,298,596	$5,317,889
2004	49.3	24.9	$1,404,856	$2,846,788
2003	17.9	45.0	$1,302,302	$7,260,488
2002	46.0	29.5	$1,433,683	$3,116,415
2001	17.6	66.6	$875,000	$4,984,784
2000	60.9	35.1	$1,176,894	$1,931,070
Washington Post
2012	45.6	0.0	$400,000	$877,724
2011	38.8	0.0	$400,000	$1,031,903
2010	93.2	0.0	$400,000	$429,070
2009	84.6	0.0	$400,000	$472,997
2008	49.3	0.0	$400,000	$811,960
2007	97.2	0.0	$400,000	$411,700
2006	46.2	6.2	$400,000	$865,708
2005	49.3	49.3	$400,000	$811,316
2004	97.3	0.0	$400,000	$410,918
2003	49.3	49.3	$400,000	$810,658
2002	63.8	0.0	$400,000	$626,771
2001	49.5	49.5	$399,996	$808,836
2000	68.6	29.9	$399,996	$583,454
Lee Enterprises
2012	43.0	31.1	$900,000	$2,093,267
2011	70.3	28.6	$809,135	$1,150,722
2010	100.0	0.0	$833,654	$833,654
2009	63.0	33.3	$833,654	$1,323,098
2008	33.4	57.2	$850,000	$2,543,453
2007	21.8	62.3	$825,000	$3,791,282
2006	31.7	59.1	$1,100,000	$3,475,229
2005	32.8	61.8	$1,875,000	$5,716,340
2004	36.7	57.4	$1,830,000	$4,983,848
2003	39.6	54.2	$1,650,000	$4,166,143
2002	40.7	53.1	$1,325,000	$3,256,632
2001	34.8	57.5	$875,000	$2,510,935
2000	58.1	17.9	$598,100	$1,029,226
New York Times
2012	16.1	10.1	$1,107,904	$6,896,611
2011	8.9	13.9	$1,000,000	$11,252,720
2010	18.9	20.0	$1,000,000	$5,279,209
2009	15.4	25.8	$962,500	$6,262,755
2008	18.6	54.4	$1,035,000	$5,578,451
2007	24.1	17.6	$1,000,000	$4,142,410
2006	22.7	53.8	$1,000,000	$4,402,678
2005	31.1	64.8	$1,377,900	$4,430,927
2004	67.7	39.7	$1,919,544	$2,835,255
2003	60.4	66.0	$1,738,265	$2,880,089
2002	43.0	65.6	$2,489,594	$5,786,891
2001	16.1	74.1	$920,000	$5,713,404
2000	38.6	49.0	$2,175,000	$5,633,714

Note. CEO = chief executive officer.

However, newspaper company CEOs earned significantly less than the average CEO of a U.S. Standard & Poor’s 500 company. In 2012, the average salary for a CEO of an S&P 500 company was US$12.3 million, nearly triple what newspaper company CEOs earned. Newspaper companies’ CEOs do much better when compared with CEOs in their business category. The average salary of CEOs in the printing and publishing industries was US$3.8 million (AFL-CIO, 2013). The CEOs of newspaper companies earned 15.8% more than the average CEO in their industry. If the Post is removed, the CEOs earned 31.6% more than the average CEO in their industry group.

All of the CEOs, with the exception of the Washington Post and Media General, earned more than the average for their industry group. Gannett’s CEO earned 124% more than the industry’s average CEO; McClatchy’s CEO earned 18% more than the average CEO; the New York Times’s CEO earned 82% more; and Lee Enterprises’s CEO earned 21% more. Only Media General, 16% less, and the Washington Post, 76% less, had CEOs who earned less than the average in their industry group.

In 2012, the average worker’s annual salary was US$34,645; CEOs, on average, earned 354 times that amount (AFL-CIO, 2013). Newspaper CEOs, on average, earned 127 times what the average worker earned.

The U.S. Bureau of Labor Statistics reported that the average salary for reporters, correspondents, and broadcast news analysts in the United States was US$37,090 in 2012. Newspaper company CEOs, on average, earned 103 times more than the average journalist. Journalists’ salaries at individual companies in this study are not publicly available.

Comparing CEO compensation for individual companies with the average journalist’s salary, Gannett’s CEO earned 229 times more than the average journalist; McClatchy’s CEO earned 121 times more than the average journalist; Media General’s CEO, 86 times more than the average journalist; the New York Times’s CEO, 186 times more than the average journalist; and Lee Enterprises’s CEO, 124 times. The Washington Post’s CEO had the lowest multiple, earning just 24 times the average journalist.

In short, despite declining revenues and declining stock prices, newspaper company CEOs on average earned significantly more than their peers in the same industry. They also earned a large multiple of the average worker’s salary, but the multiple is significantly lower than the average CEO at all S&P 500 companies.

Sources of Newspaper Company CEO Compensation

The guiding business theory that describes the relationship between companies and their managers is agency theory (Eisenhardt, 1989; Pepper & Gore, 2015). Agency theory states that the shareholders—the owners of a company—are the final authority in the company. However, owners are unable to manage the daily operations of the company, so they must hire agents—or manager—to run it. Agency theory presumes that managers favor their self-interests over the shareholders’ interests, which results in costs to the shareholders (Rosenberg, 2003). For example, managers may focus on their salaries and perks or empire building at the expense of maximizing the firm’s financial performance. To maximize shareholder benefits, managers’ and shareholders’ interests need to be aligned. The key method for accomplishing this uses the executive-compensation package. Compensation is tied to the shareholders’ interest by rewarding managers for increasing the company’s value or profits. This incentivizes managers to act in the shareholders’ best interests and, by doing so, in their own best interests (Downs, 2012).

Agency theory is relevant to the long-standing debate about how newspaper owners balance financial performance and high-quality news coverage. Newspapers sell stock to the public to raise capital, “the fundamental, underlying asset that allows firms to operate, develop, and grow” (Picard & van Weezel, 2008, p. 23). Stockholders expect the newspaper firm to maximize the profitability of the stockholders’ investment, creating an “incentive. . .to ensure that a firm’s management operates in the interest of shareholders” (An, Jin, & Simon, 2006, p. 122). Some newspaper researchers argue that publicly owned companies might therefore emphasize profits at the expense of quality news coverage that serves public and community interests (Picard, 2004; Picard & van Weezel, 2008).

However, even scholars who are skeptical about public ownership recognize that “stable, financially strong, and well-managed companies are more likely to perform well in public-interest terms because papers need financial strength to invest in serious coverage and to challenge entrenched interests such as government or corporations” (Picard & van Weezel, 2008, p. 29). This study is focused on this fundamental necessity for financial strength and how newspaper companies responded when their finances were dangerously weakened.

The need to balance profits and product quality is not unique to newspapers. All publicly traded companies must produce products and/or services that their customers value or risk financial decline. Extensive research on agency theory using many different financial measures to determine how companies’ financial goals correlate with executive compensation has had mixed results. Some research finds a correlation between financial measures and executive compensation; other research does not, leaving executive compensation as a “black box” (Bhagat & Romano, 2009; Downs, 2012).

For example, a study of CEO compensation at media companies—nearly all in the entertainment business—found no correlation between key financial measures of company performance and CEO compensation (Sigler, 2011). Non-financial measures such as the length of time as CEO and the size of the company correlated best with executive compensation. The researcher concluded entertainment media that employ highly paid entertainers might “not link CEO pay to firm performance because of the star effect where firms structure CEO compensation as it is not less than that of the talent employed by the company” (Sigler, 2011, p. 17).

Shao (2010) studied 75 publicly owned media companies to compare compensation tied to financial performance with compensation that was fixed regardless of financial performance. Shao found a higher percentage of fixed compensation for CEOs and members of the board of directors was associated with better financial performance.

A study (An et al., 2006) of 12 publicly owned newspaper companies from 1998 to 2000 found mixed support for agency theory. An increase in ownership of company stock by institutions such as banks, pension funds, or hedge funds was associated with lower profits, contradicting agency theory. But an increase in stock ownership by managers inside the company was associated with less debt, which improves a company’s financial health (An et al., 2006).

However, most U.S. companies tie the lion’s share of CEO compensation to company financial performance. In 2012, the CEOs of S&P 500 companies earned, on average, US$1.1 million in salary and US$0.3 million in bonuses. The majority of CEO compensation came from grants of stock, stock options, non-equity incentive plans, and deferred compensation. On average, only about 11.1% of CEO compensation comes from salary and bonuses. The largest share of compensation, 36.6%, comes from stock grants. Stock options account for 17.9% of average CEO compensation, 21.1% from non-equity incentive plans, 12.2% from pension and deferred compensation, and 4% from other sources (AFL-CIO, 2013). More than half of CEO compensation is tied directly to stock price, which should incentivize CEOs to maximize shareholder value, and is consistent with agency theory.

However, Table 2 shows the makeup of CEO compensation in the newspaper industry was moving in the opposite direction. Newspaper CEOs receive a larger percentage of their compensation from salary and bonuses than the average U.S. CEO. The 2012 range in cash compensation as a percentage of total compensation, however, is large: Gannett CEO, 29.4%; Media General CEO, 29.0%; Lee Enterprises CEO, 43.0%; New York Times CEO, 16.1%; Washington Post CEO, 45.6%; and McClatchy CEO, 10.9%.

In recent years, there has been a decline in the percentage of compensation that newspaper company CEOs receive from stock and stock options. This may result from the precarious pricing of their companies’ stock. Many stock options granted to newspaper company CEOs in the early and mid-2000s had strike prices that were never reached, making these options worthless. More likely, newspaper company CEOs may be demanding more of their compensation in cash that is paid immediately instead of a less-than-certain future payout based on precarious stock prices. Not tying a large amount of CEO compensation to stock prices and other long-range financial goals means newspaper company CEOs may be incentivized to focus on short-term goals at the expense of long-term shareholder value.

Changes in Compensation of Newspaper Companies’ CEOs and Top Management

Table 3 reports total compensation of the newspaper companies’ five highest paid executives, the percentage of compensation from salary and bonuses, and the average compensation for the four executives who are not the CEO. Compensation for the five highest earning executives reached a high of US$81.1 million in 2006 and a low of US$50.7 million in 2008, a decrease of 38%. Compensation of the companies’ top executives has increased steadily since 2008. By 2012, the top executives earned US$71.9 million, an increase of 41.8% since 2008. Except for 2006, this amount is more than the top executives earned in any year since 2000, despite their companies suffering large revenue declines and plummeting market values.

Table 3

Compensation of Five Highest Paid Executives

	Cash compensation (US$)	Non-cash (US$)	Total (US$)	% cash	Average compensation without CEO (US$)
Gannett
2012	$6,047,693	$11,534,874	$17,582,567	34.4	$2,282,242
2011	$6,692,020	$16,838,159	$23,530,179	28.4	$2,818,766
2010	$7,329,488	$18,788,433	$26,117,921	28.1	$4,178,218
2009	$6,620,160	$7,175,475	$13,795,635	48.0	$2,274,236
2008	$5,544,667	$3,818,823	$9,363,490	59.2	$1,557,005
2007	$7,523,175	$10,044,719	$17,567,894	42.8	$2,505,296
2006	$7,696,426	$20,016,327	$27,712,753	27.8	$4,890,577
2005	$8,544,875	$13,563,509	$22,108,384	38.6	$3,440,691
2004	$8,150,780	$12,102,138	$20,252,918	40.2	$2,146,676
2003	$7,490,236	$16,057,455	$23,547,691	31.8	$2,667,005
2002	$8,010,000	$16,792,428	$24,802,428	32.3	$3,056,629
2001	$7,650,000	$15,470,139	$23,120,139	33.1	$2,724,578
2000	$7,549,136	$12,702,240	$20,251,376	37.3	$2,494,927
McClatchy
2012	$2,747,545	$7,506,648	$10,254,193	26.8	$1,440,646
2011	$2,690,539	$6,746,929	$9,437,468	28.5	$1,270,959
2010	$2,882,000	$4,668,085	$7,550,085	38.2	$961,139
2009	$2,864,904	$5,805,583	$8,670,487	33.0	$1,229,315
2008	$3,400,000	$2,988,035	$6,388,035	53.2	$905,182
2007	$3,365,000	$5,487,538	$8,852,538	38.0	$1,058,593
2006	$4,605,000	$6,098,365	$10,703,365	43.0	$1,275,389
2005	$4,480,000	$6,597,380	$11,077,380	40.4	$1,887,157
2004	$4,695,000	$2,220,993	$6,915,993	67.9	$1,099,007
2003	$4,105,000	$3,122,637	$7,227,637	56.8	$1,023,488
2002	$3,909,712	$3,254,194	$7,163,906	54.6	$1,020,188
2001	$2,907,741	$3,751,256	$6,658,997	43.7	$984,268
2000	$2,914,412	$1,925,728	$4,840,140	60.2	$672,971
Media General
2012	$2,468,033	$6,281,270	$8,749,303	28.2	$1,378,190
2011	$2,443,994	$2,791,648	$5,235,642	46.7	$949,276
2010	$2,767,800	$3,456,191	$6,223,991	44.5	$1,059,320
2009	$2,970,126	$2,325,666	$5,295,792	56.1	$986,386
2008	$2,933,500	$4,223,753	$7,157,253	41.0	$1,097,347
2007	$3,075,000	$5,602,142	$8,677,142	35.4	$1,668,752
2006	$3,030,000	$9,048,646	$12,078,646	25.1	$2,228,356
2005	$4,427,556	$11,031,412	$15,458,968	28.6	$2,535,270
2004	$4,501,425	$3,041,202	$7,542,627	59.7	$1,173,960
2003	$3,957,322	$11,728,931	$15,686,253	25.2	$2,106,441
2002	$3,939,575	$3,063,709	$7,003,284	56.3	$971,717
2001	$2,508,062	$7,083,186	$9,591,248	26.1	$1,151,616
2000	$3,241,425	$1,164,249	$4,405,674	73.6	$618,651
Washington Post
2012	$2,876,750	$10,162,846	$13,039,596	22.1	$3,040,468
2011	$2,945,000	$8,103,994	$11,048,994	26.7	$2,504,273
2010	$2,792,300	$5,369,342	$8,161,642	34.2	$1,933,143
2009	$2,688,300	$6,452,411	$9,140,711	29.4	$2,166,929
2008	$2,440,000	$3,285,996	$5,725,996	42.6	$1,228,509
2007	$2,120,000	$6,963,456	$9,083,456	23.3	$2,167,939
2006	$2,868,165	$1,865,971	$4,734,136	60.6	$967,107
2005	$2,869,494	$1,318,675	$4,188,169	68.5	$844,213
2004	$2,869,494	$1,318,675	$4,188,169	68.5	$944,313
2003	$2,704,494	$1,843,992	$4,548,486	59.5	$934,457
2002	$2,521,792	$733,277	$3,255,069	77.5	$657,075
2001	$1,666,484	$1,520,911	$3,187,395	52.3	$594,640
2000	$2,189,827	$2,056,969	$4,246,796	51.6	$915,836
Lee Enterprises
2012	$2,449,000	$2,118,882	$4,567,882	53.6	$618,654
2011	$2,227,328	$707,559	$2,934,887	75.9	$446,041
2010	$2,251,847	$0	$2,251,847	100.0	$354,548
2009	$2,251,847	$1,156,732	$3,408,579	66.1	$521,370
2008	$2,296,000	$3,560,388	$5,856,388	39.2	$828,234
2007	$2,185,000	$4,898,432	$7,083,432	30.8	$823,038
2006	$2,686,500	$4,070,345	$6,756,845	39.8	$820,404
2005	$3,624,400	$6,057,929	$9,682,329	37.4	$991,497
2004	$4,006,700	$4,321,956	$8,328,656	48.1	$836,202
2003	$3,549,700	$3,924,516	$7,474,216	47.5	$827,018
2002	$3,239,100	$3,010,792	$6,249,892	51.8	$748,315
2001	$2,997,900	$3,034,692	$6,032,592	49.7	$880,414
2000	$3,130,600	$1,458,112	$4,588,712	68.2	$889,872
New York Times
2012	$3,115,899	$14,501,347	$17,617,246	17.7	$2,680,159
2011	$3,797,845	$19,905,594	$23,703,439	16.0	$3,112,680
2010	$3,750,434	$13,883,860	$17,634,294	21.3	$3,088,771
2009	$4,058,863	$14,158,782	$18,217,645	22.3	$2,988,723
2008	$3,849,778	$9,836,371	$13,686,149	28.1	$2,026,925
2007	$4,216,500	$8,925,350	$13,141,850	32.1	$2,249,860
2006	$3,850,000	$10,517,806	$14,367,806	26.8	$2,491,282
2005	$3,660,119	$13,273,640	$16,933,759	21.6	$3,125,708
2004	$7,008,617	$3,946,527	$10,955,144	64.0	$2,029,972
2003	$6,206,278	$5,842,217	$12,048,495	51.5	$2,292,101
2002	$8,708,952	$13,942,859	$22,651,811	38.4	$4,216,230
2001	$3,922,833	$10,936,557	$14,859,390	26.4	$2,286,496
2000	$7,950,000	$9,777,128	$17,727,128	44.8	$3,023,353

Note. CEO = chief executive officer.

Examining the companies individually, the overall compensation of the five highest paid executives was not markedly different in 2012 from their highest earning year. Since 2010, Gannett’s top executives have been earning nearly the same as they earned in 2006, their highest earning year. This is also the case with McClatchy and The New York Times. The Washington Post’s top executives were better compensated in 2012 than in any year of this century. Only Lee Enterprises’s top executives saw a significant drop in their compensation. In 2012, total compensation of Lee Enterprises’s top five executives was US$4.6 million compared with US$9.7 million in 2005. Overall, the compensation levels of the top executives of newspaper companies seems largely unaffected by the financial collapse of their companies.

As with CEOs, the compensation of the top executives was shifting away from stock and stock options to cash. Comparing salary and bonuses with compensation from other sources, including stock, stock options, non-equity incentives, and pension and deferred compensation, shows the largest share of the top executives’ compensation comes from salary and bonuses. Compensation from salaries and bonuses is substantial but varying: Gannett, 34.4%; McClatchy, 26.8%; Media General, 28.2%; Washington Post, 21.1%; Lee Enterprises, 53.6%; and New York Times, 17.7%. Salary, and especially bonuses, are tied to a company’s short-term financial performance, so these executives may be incentivized to focus on those goals instead of long increases in stockholder value.

If the compensation of the top executives of the newspaper companies has recovered from all-time lows, the CEOs of these companies have done even better. The combined compensation of the CEOs of the six companies was US$38.3 million in 2005, the highest of any year. The combined compensation of the CEOs hit its low in 2008 at US$21.4 million and has increased steadily since then. In 2011, CEO compensation was 56.5% above 2008 levels, and in 2012, it was 33.6% higher at US$28.6 million compared with 2008.

Examining the companies individually, Gannett’s CEO earned a low of US$3.2 million in 2008; by 2012, his compensation had increased to US$8.5 million, a 165.6% increase in 4 years. McClatchy’s CEO earned US$10.3 million in 2012, only 8.4% less than his all-time high compensation of US$11.1 million in 2005, this despite the company losing much of its market value, writing off large amounts of goodwill, and being warned it might be delisted by the New York Stock Exchange. Both the New York Times’s and the Washington Post’s CEOs earned more in 2012 than in any year of this century. Only the CEOs of Lee Enterprises and Media General saw significant declines in their compensation after 2000. Lee Enterprises’s CEO’s best earning year was 2005 when she earned US$9.7 million in total compensation; in 2012, her compensation was US$4.6 million, a drop of 52.6%. But her total compensation has doubled since 2010 from US$2.3 million to US$4.6 million. Media General’s CEO’s compensation was US$15.5 million in 2005; by 2012, his compensation had dropped 43.2% to US$8.8 million. Still, his 2012 compensation is still 69.2% more than it was in 2011.

In summary, the CEOs and top executives of the newspaper companies are earning less than they did in their very best years, but their compensation showed steady large increases over the most recent 2 years in this study. Compensation is now close to what they were earning in their best years, and in some cases, they earn more than at any time this century. Despite significant drops in company revenues and continued declines in market value, CEO compensation has increased in the study’s last few years.

CEO Compensation Compared With Market Value

This section asks if CEO compensation tracked the changes in their companies’ market value. The market value of the six newspaper companies hit its all-time high of US$37.9 billion in 2004. Market value then went into a free fall, declining to US$4.7 billion in 2009, a drop of 87.6%. Within 6 years, shareholders lost a staggering US$33.2 billion. The market value increased to US$9.8 billion in 2010, but dropped 31.6% to US$6.7 billion in 2012. With the exception of 2010, each year since 2004 has seen a decrease in the companies’ market value.

Figure 1 compares CEO compensation with the companies’ market value. As dollar-to-dollar comparisons would skew the results, the figure uses percentages. Each year’s percentage is based on the difference between the current year and the year of the highest market value and highest CEO compensation. For market value, 2004 is the base year or 100%; for CEO compensation, 2005 is the base year or 100%. Changes in CEO compensation followed changes in market value until 2009. Declines in market value did not recover after 2009. However, the percentage declines in CEO compensation after 2009 were much less than the percentage declines of market value. In 2009, the trend lines decoupled and begin moving in opposite directions.

Figure 1

Market Value Compared With CEO Compensation

This decoupling of market value and compensation should not be surprising considering the shift in CEO compensation away from stock and stock options and toward salary and bonuses. This change provides less incentive for CEOs to focus on increasing shareholder value.

CEO Compensation Compared With Employment

All of these newspaper companies shed significant numbers of employees following the financial collapse that started in 2005. Figure 2 compares CEO compensation with employment at the companies in this study. As with Figure 1, the comparison is based on the percentage difference in employment from the year of the highest employment, 2006, and the percentage difference in compensation from the year of the highest CEO compensation, 2005. Media General did not report employment figures for every year and is not included in Figure 2.

Figure 2

Employment Compared With CEO Compensation

The percentage changes in employment and compensation closely follow each other until 2009. As employment dropped from its peak in 2006, CEO compensation also declined. But, in 2009, the trends decoupled. Changes in employment remained relatively steady. However, changes in CEO compensation moderated and trended upward. This indicates that employment does seem to be a factor in setting CEO compensation. (Research shows that CEO turnover significantly increases when companies lay off large numbers of employees [Billger & Hallock, 2005]. However, this did not occur in the newspaper industry [Soloski, 2015].)

A Note on the Data and the Newspaper Industry

Privately owned newspaper companies are not required to release to the public financial data. However, publicly owned newspaper companies sell stock to the public and must therefore release detailed financial data to comply with federal laws and regulations. This study uses data released by the six public companies that (a) were primarily in the newspaper business with consistent structure and ownership throughout the study and (b) were publicly traded continuously from 2000 to 2013, which meant they were required to file financial reports with the SEC. (News Corp. is not included because its newspapers represent a small part of the company and its revenues. A. H. Belo and the E. W. Scripps Company spun off their newspapers into separate companies in 2007. Tribune Co. filed for bankruptcy in 2008 and came out of it in 2012 as a privately held company. In 2013, Tribune Co. spun off its newspapers into a separate company. Community Newspaper Company Holdings, Inc., now CNHI, is a private company and a subsidiary of the Retirement Systems of Alabama. None met the criteria to be included in this study.)

Relationships of CEO Compensation and Financial Measures

The descriptive findings in this study and Sigler’s (2011) findings of no relationship between financial variables and CEO compensation at mostly media entertainment firms suggest a more rigorous test of relationships between newspaper executive compensation and company financial performance. The test uses three key measures: return on equity (ROE), return on assets (ROA), and market value. ROE is the return on stockholders’ equity and measures a company’s efficiency at earning profits on the capital and retained earnings that shareholders provide. ROEs between 15% and 20% are considered good. ROA shows how profitable a company’s assets are when used to generate revenue. ROA is a measure of management efficiency in employing company assets. It is difficult to compare ROA across industries; however, an ROA above 5% is considered good. Finally, again, market value is the number of shares times the stock price. Stock price is a function of investor demand and is assumed to be driven by a company’s financial performance.

Table 4 provides descriptive statistics for 14 years, 2000 to 2013, of CEO Salaries and the market value, ROE, and ROA from six newspaper companies in this study. As noted earlier, an extra year of data was added to increase the sample size of the regressions. The data summarized in Table 4 are cross-sectional time-series, also called panel data. Each company represents a panel. Each company’s annual financial data represent variables that change over time but do not change across companies. Regressions using panel data control for unobserved variables, such as different company practices, while measuring individual heterogeneity within each company.

Table 4

Descriptive Statistics 2000-2013 (n = 14 for Each Company)

Company	CEO salary (US$)	Market value (US$)	ROA (%)	ROE (%)
Gannett
M	$9,251,389	$11,706,140,413	1.11	−20.55
SD	$2,967,195	$7,874,872,154	24.94	145.37
Minimum	$3,135,469	$874,868,691	−85.26	−525.24
Maximum	$12,879,673	$22,717,076,274	13.24	33.68
Range	$9,744,204	$21,842,207,583	98.50	558.92
Lee Enterprises
M	$2,746,369	$878,474,903	−1.36	13.15
SD	$1,539,218	$779,935,588	16.04	22.89
Minimum	$833,654	$12,576,000	−43.19	00.00
Maximum	$5,716,340	$1,995,243,000	31.41	80.86
Range	$4,882,686	$1,982,667,000	74.60	80.86
McClatchy
M	$3,408,743	$881,737,077	−1.86	−36.52
SD	$1,125,531	$708,596,107	18.78	174.91
Minimum	$1,295,386	$28,100,000	−66.12	−642.95
Maximum	$5,601,811	$2,365,373,000	8.01	31.78
Range	$4,306,425	$2,337,273,000	74.13	674.73
Media General
M	$3,046,785	$733,232,078	−3.77	−46.16
SD	$1,731,909	$545,540,783	13.47	119.43
Minimum	$1,250,296	$40,526,000	−47.36	−407.05
Maximum	$7,260,488	$1,443,600,000	7.44	12.53
Range	$6,010,192	$1,403,074,000	54.80	419.58
New York Times
M	$5,405,038	$3,318,928,571	3.94	10.58
SD	$2,051,457	$2,065,466,252	6.61	25.29
Minimum	$2,835,255	$734,000,000	−14.09	−66.29
Maximum	$11,252,720	$6,380,000,000	12.93	38.68
Range	$8,417,466	$5,646,000,000	27.03	104.97
Washington Post
M	$668,254	$3,426,857,142	4.67	8.25
SD	$215,398	$920,194,950	2.02	4.32
Minimum	$402,550	$2,100,000,000	1.27	0.72
Maximum	$1,031,903	$4,927,000,000	7.71	13.79
Range	$629,353	$2,827,000,000	6.43	13.07

Note. CEO = chief executive officer; ROA = return on assets; ROE = return on equity.

A fixed-effects regression model was used to control for the time invariant characteristics of each company. Controlling for characteristics associated with each company allows us to assess the net effect of the independent variables on CEO Salary. The independent variables are each company’s market value, ROA, and ROE.

A general model of panel data based on Drukker (2012) is

y_{i t} = α + X_{i t} β + z_{i} λ + ν_{i} + ε_{i t},

where y_it is the dependent variable for company i at time t and x_it is a vector of independent financial variables that vary by company i and time t. In this model, z_i is the vector of time invariant variables represented by the companies; v_i is an unobserved, time-invariant panel effect; and ε_it is the error term representing idiosyncratic shocks that vary with companies and time. The regression coefficients are represented by β and λ.

Panel data violate regression assumptions for ordinary least squares (OLS), such as independence of the error term (Drukker, 2012). In a fixed-effects regression, v_i may be related to x_it or z_i. Alternatively, v_i might be directly correlated with the error term, producing what is called a random effect. (Random effects are associated with individual variables. Random effects exist if variation across companies is not systematically associated with CEO Salary. The models in this study were tested with random effects, and none were found.) Panel data should therefore be analyzed using statistical software that accounts for the properties of panel data.

The MIXED procedure in SPSS was developed for panel data. The MIXED models provide tools for estimating both fixed and random effects while adjusting for issues associated with panel data (SPSS, 2005). MIXED uses restricted maximum likelihood estimation (RMLE) to estimate parameters in panel-data regressions. This technique estimates the coefficients that are most likely to be correct given the data in the regression.

Companies in this study are represented in six panels, and each panel has data for 14 years. Therefore, the panels are balanced. Table 4 provides descriptive statistics for each panel. Panel regressions assume the data are drawn from sampling distributions where the means of the variables are normally distributed (SPSS, 2005). It follows from the Central Limit Theorem that if the samples are large enough, “the sampling distribution of means are normally distributed regardless of the (actual) distribution of the variables” (Tabachnick & Fidell, 1996, p. 71). However, outliers are still an issue in panel data.

The sample for each panel in this study is relatively small, so data for all six panels were aggregated and examined for normality and the presence of outliers. Data were also examined panel by panel. Data for all four variables—market value, ROA, ROE, and CEO Salary—were not normally distributed. However, the degree to which the data violated assumptions of normality varied. The violations of normality were not consistent from panel to panel or from variable to variable.

All four variables were transformed using a natural log. (ROA and ROE have negative values. Constants were added to each variable so the minimum value = 1 before the transformation to a natural log.) Then, OLS regressions were conducted with diagnostics for multicollinearity because Table 5 shows that ROE and ROA for the aggregate data are strongly correlated (R = .87, p < .005, n = 84). Multicollinearity occurred in OLS regressions that included the natural log of the independent variables: ROE, ROA, and market value. However, multicollinearity was not present when the untransformed independent variables were included with the natural log of the dependent variable, CEO Salary.

Table 5

Pearson’s R (n = 84)

	CEO salary	Market value	ROA	ROE
CEO Salary	—
Market value	.654**	—
ROA	.110	.192	—
ROE	.067	.088	.866**	—

Note. CEO = chief executive officer; ROA = return on assets; ROE = return on equity.

p ≤ 05. **p ≤ .005.

The fixed-effects regressions in this study therefore used the natural log of CEO Salary as the dependent variable. These regressions used the untransformed independent variables. The log of salary improved the model fit compared with fixed-effects regressions that used untransformed salary as the dependent variable.

The MIXED models also handle two other issues in the data for this study. MIXED models can handle serial correlation, also called autocorrelation, which exists if error terms from different time periods are correlated. This means a variable’s value in a current time period is correlated with its value in previous time periods. An SPSS test using the Box–Ljung statistic was significant (p < .005) for CEO Salary and market value, indicating both variables have serial correlation. (The stationarity properties of a time series do not need to be addressed in panel-data regressions [Drukker, 2012]). MIXED allows regressions to be modeled using a first-order autoregressive (AR1) procedure. In an AR1 model, the current value of a variable is a linear function of the variable’s value in the previous time period plus a disturbance term. Regressions using the ARI procedure produced results that were identical to results that did not use this procedure. Therefore, the AR1 regressions were not included in this study.

A check for heteroskedasticity was conducted using the General Linear Model (GLM) procedure in SPSS. The GLM regression used a fixed-effects model with the natural log of CEO Salary as the dependent variable and market value, ROE, and ROA as the independent variables. Levene’s Test for Equality of Variances was significant (p = .05), so the null hypothesis of equal variances was rejected. Unlike GLM, the MIXED procedures “handle. . . unequal variances” (SPSS, 2005, p. 1). Finally, a check of the residuals from the MIXED regressions using a P-P plot showed they were normally distributed.

Regression Results

Results in Table 6 are for fixed-effects regressions using two models, one with Gannett included and one without Gannett. This was done because the analysis that included Gannett showed a statistically significant, or non-random, relationship between CEO pay and a company’s market value. This relationship was not statistically significant when Gannett was dropped from the analysis, suggesting Gannett may be the only company that ties CEO pay to the financial measures in this study.

Table 6

Estimates of Fixed Effects (DV = CEO Salary)

Parameter	Model 1(n = 84)	95% CI lower bound	95% CI upper bound	Model 2(n = 70)	95% CI lower bound	95% CI upper bound
Intercept	13.27** (0.13)	13.02	13.53	13.24** (0.21)	12.82	13.66
Market value	2.99E-11* (1.52E-11)	−5.05E-13	6.02E-11	5.01E-11(5.15E-11)	−5.29E-11	1.53E-10
ROA	−0.50(0.67)	−1.85	0.84	−1.30(0.78)	−2.86	0.25
ROE	0.096(0.097)	−0.098	0.291	0.11(0.10)	−0.09	0.32
Gannett	2.38** (0.21)	1.96	2.80	—	—	—
McClatchy	1.71** (0.17)	1.37	2.05	1.72** (0.22)	1.29	2.15
Media General	1.52** (0.17)	1.18	1.87	1.52** (0.22)	1.08	1.96
NYT	2.08** (0.17)	1.75	2.41	2.08** (0.17)	1.73	2.42
Lee Enterprises	1.34** (0.17)	0.99	1.69	1.35** (0.21)	0.92	1.77
Washington Post	0^a	—	—	0^a	—	—
Model fit
–2 restricted log likelihood	158.03			140.24
Wald Z	6.12**			5.57**

Note. CEO = chief executive officer; CI = confidence interval; ROA = return on assets; ROE = return on equity; NYT = New York Times.

Parameter set to zero because it is redundant.

p ≤ 05. **p ≤ .005.

There are two model fit statistics reported in Table 6. The log likelihoods are used to compare each model’s fit. A likelihood ratio test statistic is calculated as the difference between log likelihoods for each model (SPSS, 2005). This statistic has a chi-square distribution with degrees of freedom determined by the difference in the number of parameters for each model. The difference in likelihoods was not significant (17.79, df = 14, p = .21). The model without Gannett is not a significant improvement over the model with Gannett.

The second model fit statistic is a Wald test. The Wald test examines the null hypothesis that all coefficients in the model are zero. Table 6 shows the test is significant for both models, so we can reject the null. Model 1 in Table 6 includes Gannett in the regression. Only one financial variable, market value, is significantly associated with CEO Salary. Because CEO pay was transformed to a natural log, regression coefficients can be interpreted to mean a one-unit change in the independent variable results in a 100*β percent change in pay (Wooldridge, 2009). This means a US$1 change in market value results in an average 2.99E-9, or 0.00000000299, percent change in CEO pay. In other words, there is an average 2.99 percent change in pay for every US$1 billion change in market value.

The Model 1 regression includes a categorical variable with six levels, one for each company, to measure fixed effects from the companies. MIXED automatically set the level for Washington Post to zero. This makes Washington Post a reference point for changes in CEO pay. All remaining company coefficients are significantly associated with CEO pay. For example, Gannett is associated with a 238% average increase in pay compared with the Washington Post. Lee Enterprises is associated with a 134% average increase in pay compared with the Washington Post. The Post had the lowest average CEO Salary, which is why the other company coefficients are positive.

The Model 2 regression in Table 6 did not include Gannett. In this model, market value is not significantly associated with CEO Salary. The fixed-effects from the other companies remain significant, and the coefficients are almost identical to Model 1. This is consistent with the model of company effects as time-invariant.

The fixed-effect regressions show only a small relationship between one financial variable, market value, and CEO pay. This relationship appears to be strongly influenced by just one of the six companies, Gannett. These regressions also show that salary differences are influenced by the characteristics of the companies themselves. These results are consistent with the descriptive results suggesting financial performance was not a major factor when newspaper companies determined their CEOs’ pay.

These regressions have limitations. Other variables that are not in the regressions might have helped determine pay. If that is the case, the regressions may reflect omitted variable bias. However, the relationships in Table 6 were also found in the GLM regressions used to check for heteroskedasticity. This suggests the findings in Table 6 relationships are reliable, and valid. These findings are also similar to Sigler (2011) who did not find a statistical relationship between ROA and CEO compensation.

Discussion and Conclusion

Newspaper CEO and Top Executive Compensation

Newspaper CEOs in the study receive significantly less compensation than does the average CEO of S&P 500 companies. But within their industry group, newspaper CEOs receive compensation above the average. The multiple of CEO compensation above the average worker is far less for newspaper company CEOs than for CEOs of S&P 500 companies. Newspaper company CEOs earn as much as 229 times more than what a journalist earns and as little as 24 times.

The makeup of newspaper company CEOs’ compensation is quite different from the typical CEO of an S&P 500 company. At many companies, changes in the tax law and efforts to align the interests of CEOs and shareholders have resulted in stock and stock options providing the largest share of CEO compensation. But newspaper company CEOs receive a smaller percentage of stock and stock options than is typical. Salary and bonuses account for a greater share of newspaper company CEOs’ compensation than that of the typical CEO. The trend away from stock and stock options and toward salary and bonuses means newspaper company CEOs are incentivized to focus on short-term goals at the expense of long-term increases in shareholder value.

The top executives of newspaper companies saw little change in their compensation during the industry’s financial collapse. Top executives at most companies earned about what they earned during the industry’s best year. In some cases, they earned more than in any year of the century. These executives were also receiving a larger portion of their compensation in salary and bonuses than from stock and stock options. This again suggests they are incentivized to focus on short-term financial goals rather than on long-term goals.

Suffering the Pain of Investors and Employees

As newspaper company stock prices dropped precipitously in 2005 and large layoffs occurred, CEO compensation also dropped, though not as steeply as the declines in market value and employment levels. A few of the CEOs voluntarily turned down raises and bonuses during these years. However, CEO compensation has been increasing in recent years despite continued losses in market value. Only in 2006 did the CEOs receive more total compensation than they did in 2012.

While CEO compensation and market value tended to parallel one another through most of the century, in the last 3 years of the study they decoupled. The same disconnect between market value and CEO compensation seems to have occurred with employment: While drops in employment levels and CEO compensation tended to parallel each other, in the last study years, CEO compensation has been increasing while employment levels have been either down or flat.

Three financial measures—ROE, ROA, and market value—were examined to detect relationships with newspaper company CEO compensation. However, only one of these key financial measures was correlated with compensation, and the relationship was very small. This raises the question of how boards of directors arrive at CEO compensation in the newspaper business and what financial measures they do use when setting CEO compensation.

Did newspaper company CEOs share the financial pain of their investors and employees? They did immediately following the first major downturn in their companies’ market value and after the largest decline in employment. However, as market value continued to fall and employment remained well below the level in 2006, the CEOs and other top executives enjoyed significant increases in their compensation. For the CEOs, the pain did not last long.

Top executives of public companies have a fiduciary duty to act in the best interest of stockholders. This not only means maximizing shareholder value, but also that executives should act to keep their companies alive and thriving. Agency theory argues that tying compensation to shareholder value is an effective way to maximize both financial health and company survival. This study found little or no connection between financial measures and CEO compensation during a period when newspapers suffered a major financial decline. This is consistent with Table 3 that shows CEO compensation shifting from stock to cash. Our findings suggest concerns that newspaper executives might focus on finances at the expense of news (Picard, 2004; Picard & van Weezel, 2008; Shao, 2010) were mistaken once the industry began its decline.

The decline occurred as print newspaper audiences began shifting to digital media on the Internet. Advertising began shifting away from news to new forms of media such as search engines and social media. Newspaper executives are not to blame for these changes, but they are accountable for ineffective responses to the changes. Executives might have found more effective responses to digital competition if their compensation had been closely tied to the declining financial health of the companies they managed.

Today, only two legacy newspaper companies—Lee Enterprises and McClatchy—remain largely intact and both are teetering on bankruptcy. Gone are ionic newspaper companies Washington Post, Knight Ridder, Dow Jones, and Times Mirror, as well as others such as Central Newspapers, Journal Register, and Hollinger International. Yet other companies have either exited the newspaper business and been acquired in a merger (Media General) or spun off their newspapers into a separate company (Gannett, A. H. Belo, Tribune, and E. W. Scripps).

Into the ownership void have stepped companies that have no experience in journalism. Today, the largest owner of newspapers in the United States is Digital First Media, which owns 455 newspapers. Digital First Media is a subsidiary of Alden Global Capital, which specialized in buying distressed companies and selling assets. New Media/Gatehouse, owner of 205 newspapers, is controlled by Forest Investment Group, another hedge fund. If executives of legacy newspaper companies could not be successful running their companies, how can these hedge funds whose expertise is solely financial succeed, and what price will be paid by journalism?

References

American Federation of Labor and Congress of Industrial Organizations. (2013). Executive paywatch. Retrieved from https://maritimetrades.org/afl-cio-unveils-2013-executive-paywatch/

American Federation of Labor and Congress of Industrial Organizations. (2014). Executive paywatch 2014. Retrieved from http://www.aflcio.org/Corporate-Watch/Paywatch-2014

Jin

H. S.

Simon

(2006). Ownership structure of publicly traded newspaper companies and their financial performance. Journal of Media Economics, 19, 119-136. doi:10.1207/s15327736me1902_3

Barry

M. J.

Kairis

J. C.

(2011). Shareholder rights and corporate governance in the Dodd-Frank act. Retrieved from http://www.gelaw.com/wp-content/uploads/2015/02/Shareholder-Rights-Dodd-Frank.pdf

Barthel

(2015). State of the news media 2015: Newspapers: Fact sheet. Retrieved from http://www.journalism.org/2015/04/29/newspapers-fact-sheet/

Barthel

(2018). State of the News media: Newspapers fact sheet. State of the News Media. Retrieved from http://www.journalism.org/fact-sheet/newspapers/

Bhagat

Romano

(2009). Reforming executive compensation: Focusing and committing to the long-term. Yale Journal on Regulation, 26, 359-372.

Billger

S. M.

Hallock

K. F.

(2005). Mass layoffs and CEO turnover. Industrial Relations: A Journal of Economy and Society, 44, 463-489.

Cho

Smith

M. D.

Zenter

(2016). Internet adoption and the survival of print newspapers: A country-level examination. Information Economics and Policy, 37, 13-19. doi:10.1016/j.infoecopol.2016.10.001

10.

Collis

Olson

Furey

(2010). The newspaper industry in crisis. Harvard Business School Background Note 709-463, March 2009 (revised January 2010). Available from www.hbsp.harvard.edu/educators

11.

Cranberg

Bezanson

Soloski

(2001). Taking stock: Journalism and the publicly traded newspaper company. Ames: Iowa State University Press.

12.

Dodd-Frank Wall Street Reform and Consumer Protection Act. (2010). H.R. 4173, U.S. house of representatives. Retrieved from https://www.congress.gov/111/plaws/publ203/PLAW-111publ203.pdf

13.

Downs

R. C.

(2012). Executive compensation: In a culture of greed and selfishness is there room for a theory of enough. Faulkner Law Review, 4, 35-71.

14.

Drukker

D. M.

(2012). Panel-data analysis using STATA: ICPSR Summer Program. Inter-university Consortium for Political and Social Research. University of Michigan, Ann Arbor, MI.

15.

Edmonds

Guskin

Rosenstiel

Mitchell

(2012). State of the news media 2012: Newspapers: Building digital revenues proves painfully slow. Washington, DC: Pew Research Center.

16.

Eisenhardt

K. M.

(1989). Agency theory: An assessment and review. Academy of Management Review, 14, 57-74. doi:10.5465/amr.1989.4279003

17.

Fico

Lacy

Wildman

S. S.

Baldwin

Bergan

Zube

(2013). Citizen journalism sites as information substitutes and complements for United States newspaper coverage of local governments. Digital Journalism, 1, 152-168. doi:10.1080/21670811.2012.740270

18.

Holcomb

(2015). Why a mobile news startup couldn’t survive in the mobile news world. Washington, DC: Pew Research Center. Retrieved from http://pewrsr.ch/1Nk6z87

19.

Hou

Priem

R. L.

Goranova

(2014). Does one size fit all? Investigating pay–future performance relationships over the “seasons” of CEO tenure. Journal of Management, 43, 864-891. doi:10.1177/0149206314544744

20.

Jurkowitz

(2014). State of the news media 2014: The growth in digital reporting: What it means for journalism and news consumers. Washington, DC: Pew Research Center. Retrieved from http://www.journalism.org/2014/03/26/the-growth-in-digital-reporting/

21.

Karimi

Walter

(2016). Corporate entrepreneurship, disruptive business model innovation adoption, and its performance: The case of the newspaper industry. Long Range Planning, 49, 342-360. doi:10.1016/j.lrp.2015.09.004

22.

Leff

N. M.

Olshan

Schohn

E. F.

Yaffe

J. M.

Bergmann

M. R.

Davis

K. M.

Nelson

(2015). 2015 compensation committee handbook. New York, NY: Skadden, Arps, Slate, Meagher & Flom LLP.

23.

Listing standards for compensation committees, 10C-1 C.F.R. (2012).

24.

Murphy

K. J.

(1999). Executive compensation, chapter 38. In Ashenfelter

Card

(Eds.), Handbook of labor economics (Vol. 3, Part B, pp. 2485-2563). Amsterdam, The Netherlands: Elsevier.

25.

Pepper

Gore

(2015). Behavioral agency theory: New foundations for theorizing about executive compensation. Journal of Management, 41, 1045-1068. doi:10.1177/0149206312461054

26.

Picard

R. G.

(2004). Commercialism and newspaper quality. Newspaper Research Journal, 25, 54-65. doi:10.1177/073953290402500105

27.

Picard

R. G.

van Weezel

(2008). Capital and control: Consequences of different forms of newspaper ownership. International Journal on Media Management, 10, 22-31. doi:10.1080/14241270701820473

28.

Rosenberg

R. J.

(2003). Fiduciary duties and potential liabilities of directors and officers of financially distressed corporations. Paper presented at the International Insolvency Institute: Third Annual III Conference 2003, New York, NY. Retrieved from https://www.iiiglobal.org/sites/default/files/media/Fiduciary_Duties_and_Potential_Liabilities_Rosenberg.pdf

29.

Shao

(2010). Thinking about stakeholders: Compensation arrangements of media companies and their performance. International Journal on Media Management, 12, 5-19. doi:10.1080/14241270903408812

30.

Sigler

K. J.

(2011). CEO pay and company performance in the media industry. Business Management Dynamics, 1(5), 17-23.

31.

Soloski

(2013). Collapse of the US newspaper industry: Goodwill, leverage and bankruptcy. Journalism: Theory, Practice and Criticism, 14, 309-329. doi:10.1177/1464884912472016

32.

Soloski

(2015). Stability or rigidity: Management, boards of directors and the newspaper industry’s financial collapse. International Journal on Media Management, 17, 47-66. doi:10.1080/14241277.2015.

33.

SPSS

(2005). Linear mixed-effects modeling in SPSS: An introduction to the MIXED procedure. Retrieved from http://www.spss.ch/upload/1126184451_Linear%20Mixed%20Effects%20Modeling%20in%20SPSS.pdf

34.

Tabachnick

B. G.

Fidell

L. S.

(1996). Using multivariate statistics. New York, NY: HarperCollins College.

35.

Wooldridge

J. M.

(2009). Introductory econometrics: A modern approach (4th ed.). Mason, OH: South-Western Cengage Learning.