How The SEC Got DB Plans All Wrong

Yesterday, I wrote that the SEC contributed to the downfall of defined benefit (DB) plans. I promised to explain in more detail in a future post.

Well, the future is now (I wonder if anyone else ever used that line before).

As readers of this blog well know, for about 10 years, issuers of proxies under the auspices of the Securities and Exchange Commission have been required to disclose compensation for their five highest paid employees. This compensation has been disclosed in a little beast known as the Summary Compensation Table and in more detail in several other places.

Some of the disclosure makes sense. For example, the salary that an individual actually receives is what it is. There is no disputing that. Similarly, the matching contribution that an individual receives in her 401(k) plan or in the nonqualified analog is what it is.

Some of the disclosure makes less sense. Fingers point to the defined benefit disclosures.

What the rules ask be disclosed as compensation with respect to a defined benefit plan (qualified or nonqualified) is the increase, if any, in the actuarial present value of accrued benefits from one measurement date to the next. That seems simple and reasonable enough on its face, but that is exactly where, how, and why the SEC went wrong.

Let's make up some Illustrative numbers for Well Paid Executive who we will refer to as WPE.

• Total pension accrued benefit as of 12/31/2014: $1,000,000
• Total pension accrued benefit as of 12/31/2015: $1,100,000
• Actuarial present value "factor" as of 12/31/2014 using 2014 discount rates and 2014 mortality table: 11.00
• Actuarial present value "factor" as of 12/31/2015 using 2014 discount rates and 2014 mortality table: 11.50
• Actuarial present value "factor" as of 12/31/2015 using 2015 discount rates and 2015 mortality table: 12.50

So, the actuarial present value of WPE's defined benefit as of 12/31/2014 was 11*1,000,000 or $11,000,000. 

The actuarial present value of WPE's defined benefit as of 12/31/2015 was 12.50*1,100,000 or $13,750,000.

The compensation due to defined benefits that must be reported for WPE was 13,750,000 - 11,000,000 or $2,750,000.

But, this is not an apples to apples comparison. Really Big Company (RBC) did not actually compensate WPE for the fact that interest rates on bonds declined during the year. RBC also did not actually compensate WPE for the fact that the Society of Actuaries had finished a new mortality study. 

The actuarial present value of WPE's defined benefit as of 12/31/2015 using 2014 actuarial assumptions was 11.50*1,100,000 or $12,650,000. This means that the increase, on an apples to apples basis, in the actuarial present value of WPE's defined benefits was 12,650,000 - 11,000,000 or 1,650,000. The other 1,100,000 was due to changes in actuarial assumptions and DOES NOT REFLECT THE AMOUNT THAT RBC PAID WPE DURING 2015.

The SEC's methodology has thrown proxy disclosures out of whack. In light of Dodd-Frank and its Say-on-Pay requirement and Pay Ratio Disclosure requirement, the SEC methodology puts an inappropriate burden on companies sponsoring DB plans. For some, this may signal a reason for them to exit that space ... for all the wrong reasons.

Who Put the Dagger in the Heart of Defined Benefit

What happened to defined benefit (DB) plans? We, at least those of us who can remember those times, saw a rise of them from the passage of ERISA in 1974 until the Tax Reform Act of 1986 (TRA86). They were viewed as an affordable way for companies to provide an excellent retirement benefit to their employees, especially ones that showed loyalty to those companies. In fact, most current retirees who have been retired for at least 10 years retired in part because they had those DB plans. And, companies that sponsored them and did so responsibly did not go out of business because of them.

Unfortunately, the number of DB plans has dwindled significantly. There was not a single killer, though. The list is long. It includes:

• Congress
• The Pension Benefit Guaranty Corporation (PBGC)
• The global economy
• The e-commerce economy
• The temporary worker for hire economy
• The Financial Accounting Standards Board (FASB)
• The Securities and Exchange Commission (SEC)

Congress

What did Congress do that was so bad? Frankly, that crew of 535 individuals that changes in makeup from time to time did a lot. Between them, they rarely have anyone among them that knows much about DB plans. And, today, because their staffers who give them what they think is excellent guidance tend to be young, the advice that they actually get is colored by propaganda of the last 30 years or so. 

Nevertheless, Congress passed laws ... lots of them. And, lots of them included DB provision and each one was worse than the last. More than anything, though, because Congress intermingled retirement policy with tax policy so intimately, it eliminated plan sponsors' ability to fund DB plans responsibly.

Back in 1987, just one year after passage of TRA86, Congress, in its infinite wisdom, decided that the status quo, having existed for about a year, needed change. Included in that necessary change was limiting the amount that companies could contribute (and deduct on their tax returns) to DB plans. We were switched from a regime that was sound actuarially to one that never was and never will be.

Think about. Suppose you agree to pay someone an amount in the future. Shouldn't you be setting aside money regularly to pay for those future benefits? Of course, you should. And, DB funding used to be based on funding methods that allocated costs intelligently to the past, present, and future. But OBRA 87 began the path to dismantling those methods as Congress thought it better that DB deductions would be limited so that it could spend more money elsewhere. Thus was limited the ability of companies to responsibly fund their plans. And, as a result, as interest rates fell and there were a few significant downturns in equity markets, DB plans became [often] severely underfunded. Finally, the way these regimes worked, companies had no room to make deductible pension contributions in years that they could afford to, but were required to make large contributions in years that they could not afford them.

PBGC

For those that don't know, the PBGC is a corporation, run under the auspices of the Department of Labor, that was established by ERISA to protect the pensions of participants in corporate DB plans. The PBGC was and is funded by premiums paid by plan sponsors. 

While the purpose of the PBGC has always been to protect pensions, it has behaved over the last 30 years or so as if pensions should be in place to protect the PBGC. To the PBGC's policymakers, nothing would be a greater tragedy than the PBGC projecting that it would run out of money.

So, the PBGC lobbied Congress. And, with that lobbying, the PBGC got increased ability to intervene in business and in corporate transactions. It got increased premiums. It got even high premiums for underfunded DB plans, often ones to which the sponsors wanted to make contributions when business was good, but were precluded from doing so.

What happened?

The PBGC tried to commit suicide.

You see, as the PBGC helped to make DB plans less attractive for employers, employers froze entry to or terminated DB plans. This meant that there were fewer participants in DB plans on whom premiums were paid to the PBGC. Companies that couldn't terminate DB plans were the ones that had plans that were so underfunded that they didn't have enough money to fully fund them. Often, the PBGC had to assume responsibility for those plans. So, as the PBGC incurred more liabilities, it also got less premium revenue. It's awfully tough to run  a business that way.

Various Economies

As the US economy has become more global, more electronic, and more temporary, DB plans have lost some of their appeal, When a company was competing primarily with other US companies, providing excellent retirement benefits was important to attracting and retaining good employees. But, non-US companies didn't have to do that and as competition from offshore became fierce, retirement benefits became less of an issue.

And, as employers began to show less loyalty to their employees and were less worried about retaining them, retirement benefits lost some of their appeal. The prophecy was somewhat self-fulfilling; as retirement benefits lost some appeal, they continued to lose appeal.

FASB

Accounting for DB plans use to be more cash based than accrual based. In 1985, the FASB gave us Standards Number 87 and 88 that instructed companies how to account for DB plans. Previous to then, companies used Accounting Principles Bulletin (APB) 8, under which most companies expensed as part of P&L their cash contributions. The FASB saw this as incorrect.

So, we moved from a regime under which pension expense was somewhat controllable (a company by building up a surplus in good times could contribute less in bad times and thereby have some control over their pension expense) to one that had components that became quite volatile as the economy did. CFOs abhor volatility and thus had one more reason to not want DB plans.

SEC

The SEC didn't get deeply involved in pensions until recent times. But, when they did, they got them all wrong.

Primarily, the SEC's involvement with pensions has been in corporate proxies. For the last 10 or more years, companies have been required to provide as part of their proxy materials a table of Summary Annual Compensation for (generally) their top five paid employees. Included in compensation is the increase in actuarial present value from one year to the next of defined benefit pensions both qualified and nonqualified. 

Often times, that increase is largely due to a decrease in underlying discount rates or more recently, to a change in mortality assumptions. Yes, according to SEC rules, when the Society of Actuaries publishes a new, more up-to-date mortality table, that represents compensation to executives. And, of course, the media loves to look at those numbers as if someone had written checks for those amounts in those years to those executives. And, some investors look at those numbers and want to use them to explain how the companies in which they invest overpay their executives. (This is not to say that they don't overpay their executives, but when a large piece of the overpayment is simply due to year over year changes in actuarial assumptions, the logic is bad.) 

There are better ways. Actuaries understand them. And, actuaries, in some cases, have tried to explain them to the SEC. But, the SEC has chosen to listen more to attorneys, accountants, lobbyists, and unions. They carry more sway and are more emotional about this issue, but generally speaking, they don't get it.

I'll write about a better way in another post soon.

But, in the meantime, now you know where the dagger came from. DB plans use to allow people to look forward to retirement. For most, they don't anymore.

SEC Finalizes Pay Ratio Rule -- Read the Plain English Description Here

Wednesday, after much controversy over the last five years, the Securities and Exchange Commission (SEC) released its final rule under Dodd-Frank Section 953(b) sometimes known as the Pay Ratio Rule. I have friends who are executive compensation attorneys and if you need legal advice on this rule, I can recommend any number of them to you, but I am going to write about it from a practical standpoint in plain English. What happened?

First, I'd like to commend the SEC. The statute on this rule has been very controversial. The SEC, in my opinion, has taken an approach that remains largely faithful to the exact wording of the statute and fully faithful to the intent of the statute (I'm not here to argue if the statute is worthwhile) while at the same time being sensitive to the concerns of employers with regard to the potential cost of compliance. It's rare that a government agency handles such a quandary this well.

Back in 2013, the SEC released a proposed rule on this topic. Since that time, the SEC received 287,400 comments on the proposed rule. More than 285,000 of them were form letters, but that still means that roughly 2,000 people took the time to write customized comments. To the credit of the Commissioners, they appear to have considered every last one of them. What they have crafted is practical, assuming that you find the result of the work practical.

What does the rule say? Here we go.

In its definitive proxy, each registrant shall disclose three items (at least since the rule says that the disclosure may be augmented):

• The pay (as defined for proxy purposes) of the PEO (generally known as the CEO or Chief Executive Officer),
• The pay (same definition) of the median-compensated employee of the employer, and
• The ratio of the first item to the second expressed as some number (integer will work) to 1.

Identifying the median compensated employee can be a very costly process. Consider a company with 1001 employees. The median compensated will be the one whose compensation is more than that of 500 others and less than that of 500 others. In order to determine this (by the letter of the law), one would need to determine the annual total compensation (that's the proxy compensation) or ATC for each of the 1001. They would need to be ranked and then we would find the 501st person. That is a lot of work.

The final rule allows for two significant simplifications for purposes of determining the median employee:

• Companies may choose to use sampling techniques in order to reasonably determine who the median paid employee is, and
• Companies may use and consistently applied measure of compensation from payroll or tax records.

While the first of those may be more trouble than it is worth, the second should be a big help to lots of companies.

Further, once a company establishes a median employee, it may use that same employee for three years provided that there have not been significant changes (undefined term) in the compensation practices or the makeup of employees. If that employee terminates, then the company may reasonably select a similarly situated employee as a replacement.

Many commenters were concerned about the disclosures for multi-national companies especially those with significant numbers of employees in lower cost-of-living countries. Certainly, for example, $50,000 per year goes further in Kyrgyzstan than it does in the US. The final rule allows companies to adjust (on a nation-by-nation basis) compensation for cost-of-living differences.

Calculating proxy compensation can be cumbersome. It includes other than just cash compensation. So, for companies with defined benefit plans and broad-based equity compensation arrangements, it is entirely possible that multiple outside experts would need to be engaged. While that remains the case, the final rule allows companies to make reasonable estimates of components of compensation.

The statute makes clear that an employee is every employee worldwide, whether full-time, part-time, temporary, or seasonal of the controlled group. The final rule allows for all of these simplifications or adjustments:

• A determination date applied consistently within 3 months of the end of the fiscal year
• Only subsidiaries included in the consolidated financial statements need be considered
• Employees where data may be unattainable due to national (or EU) privacy rules may be excluded
• De minimis numbers of employees (up to 5% in total) may be excluded on a country by country basis

Let's look in more detail at those last two. Suppose the number of employees excluded under the privacy rule exception exceeds 5%. Then you are done with your exclusions. On the other hand, if your privacy exclusions are exactly 2% of your total population, then you may exclude other countries whose total employee population is less than an additional 3% of your total population. If, for example, you can't find another country with fewer than 3% of your total employees, then you are done with your exclusions.

In preparing these disclosures, companies will make lots of assumptions, simplifications, and estimates. All must be disclosed.

In somewhat of a gift to employers, additional disclosures and ratios are permitted, but not required so long as the additional disclosures and ratios are no more prominent than the required ones. I think this could be useful.

Consider a company with its management team and sales force in the US, but the bulk of its production facilities in third world countries (I'm not weighing in on whether this is a good or responsible practice or not). Because manual labor is particularly inexpensive in Burkina Faso, for example, Everybody's Favorite Company (EFC) has an extremely high pay ratio, say 10000 to 1. Its CEO had total compensation of $10 million and most employees in Burkina Faso earned only $1000. And further, EFC can't find cost-of-living data for Burkina Faso, so it is not able to do that adjustment. EFC is perhaps rightfully concerned about its pay ratio disclosure, so it elects to do a second pay ratio disclosure limited specifically to US employees. In this case, the ratio declines to 100 to 1.

A second company with a December 31 fiscal year end, everest.com, does a massive holiday business. As a result, Everest has a high pay ratio reflective of its hiring each year of seasonal employees. In fact, in a typical year, Everest has more than twice as many employees from September 1 through December 31 than it does the rest of the year. As a result, Everest reports a pay ratio of 750 to 1. Everest doesn't like this, so it chooses to determine an additional ratio of all but seasonal employees. The company is much more pleased to find that this ratio is only 175 to 1.

Generally, companies are required to report the pay ratio for any fiscal year beginning on or after January 1, 2017 (there are exceptions for certain new filers and emerging companies). This means that the first required disclosures (companies are encouraged to disclose before then) will generally be in the early months of 2018.

The final rule is long and complex. There are many legal issues around it and for those you should contact an attorney. 

There are also issues that are far more consultative in nature. They will generally require quantitative acumen, actuarial knowledge, and comfort with executive compensation, as well as a focus on business issues. For those, you should just click here.

Possible Delay of Pay Ratio Regulations -- Practical or Political?

Right before Thanksgiving, the House Committee on Financial Services sent a letter to the Chair of the SEC asking that guidance on the so-called Dodd-Frank Pay Ratio rule (Section 953(b)), and therefore implementation as well be delayed. The letter called this implementation a low priority. No similar correspondence has come from the Senate or any of its committees.

Gee, what a surprise that is. House committees are controlled by Republicans currently while Senate committees are in the hands of Democrats for few more weeks. Dodd-Frank was another bill and then law praised by Democrats and loathed by Republicans. The letter points out that the law does not specify a time by which the SEC must promulgate a rule on this issue. Additionally, the letter tells the SEC that the pay ratio disclosure has little value and that knowing the pay ratio for publicly traded companies back in the 2006-2008 timeframe would have done nothing to avert the financial crisis.

Let's regroup for a moment. For those who don't know, Section 953(b) of the Dodd-Frank law requires that registrants disclose a single number -- the ratio of the pay of the median-paid employee in the company to that of the CEO. Pay for this purpose is pay as defined in the executive compensation section of the proxy. I have argued that while such relative information may be useful to some that this calculation is the wrong one and needlessly burdensome. Last year, the SEC issued a proposed rule on the pay ratio calculation. While it provides a means of simplification for many companies, companies that do take advantage of the sampling techniques proposed will not find them simple by any means.

So, is the single number valuable? I don't think so. If one insists that such a comparison is valuable, then to me, it would be far more instructive to show CEO compensation and a chart of commonly held jobs within the company showing a compensation range for actual or hypothetical full-time employees(if no full-time actual employees exist for such job classifications) for each of those jobs alongside the CEO compensation.

Advocates of the pay ratio point out companies in the fast food industry. They note that CEOs make many millions of dollars while most fast food workers earn minimum wage or only slightly more. In this case, is the pay ratio valuable? To me, it's not. Running a multi-billion dollar business and defining its strategy to compete and grow has no similarities to monitoring a drive-thru window. Comparing the compensation of two such individuals makes no sense to me.

Currently, the SEC has three Democrats and two Republicans. The Chair is a Democrat. The commissioners still have meaningful Dodd-Frank work to do whether they make 953(b) a priority or not.

I don't know what they will do, but thus far, they have not shown a predisposition to act quickly on ensuring that the pay ratio is disclosed for registrants.

PCAOB Expands Purview of Accounting Profession to Review Compensation Arrangements

Unless you are in a field related to accounting or review of public companies, you may never have heard of the PCAOB, or more formally, the Public Company Accounting Oversight Board. It is actually a private-sector, non-profit company that was created by the Sarbanes-Oxley Act (SarbOx or SOx) in 2002. According to SOx, the PCAOB is to, among other things, oversee the audits of public companies. In part, to do so, the PCAOB creates a set of auditing standards that must be approved by the Securities and Exchange Commission (SEC).

Got that?

So, why am I writing about this? Earlier this year, the PCAOB issued Auditing Standard 18. In late October, the SEC approved it. And, so it is.

Auditing Standard 18 (AS18) is a lovely document. It's 223 pages of, and I can't think of a better word for it, stuff carrying the clearly far too brief title, "Related Parties and Amendments on Significant Unusual Transactions and a Company's Financial Relationships and Transactions with its Executive Officers."

Got that as well?

Among other things, AS18 asks that auditors, among other things:

1. Read the employment and compensation contracts between the company and its executive officers
2. Read the proxy statements and other relevant company filings with the SEC and other relevant regulatory agencies that relate to the company's financial relationships and transactions with its executive officers
3. Obtain an understanding of compensation arrangements with senior management other than executive officers including incentive compensation arrangements, changes or adjustments to those arrangements, and special bonuses
4. Inquire of the Chair of the Compensation Committee as well as outside compensation consultants
5. Obtain an understanding of expense reimbursement policies with respect to executive officers

All of this is to be done to identify risks of material misstatement.

I understand why all of this is being done. Early in this century, there were a number of corporate scandals resulting from apparently fraudulent misstatement of financials. Many of you remember Enron, WorldCom, and Tyco. This is intended to be one more step to lessen the likelihood of such abuses and to restore faith that corporate America, as a whole, are being good citizens.

We place lots of faith in auditors in this regard. Now, I have friends and acquaintances who are auditors. Some of them are very good and knowledgable. And, this is not intended to say that the rest (the complement of some of them) are not good, but in any profession, some practitioners will always be better than others.

Here, however, the PCAOB and the SEC are either assuming that auditors have sufficient expertise in compensation arrangements to handle this undertaking or that the firms of which they are a part have this expertise internally to which the auditors may refer. At the Big 4, this is probably the case. They have massive staffs and are able to engage specialists to handle such complex questions. How about the next tier of auditing firms? Do they have this expertise? I don't know, but I suspect the answer is sometimes. Let's take it down one more tier. Do those firms have that expertise? Again, I'm guessing, but I suspect that the answer is not very often. And, if we move to the smallest of the auditing firms, I would be inclined to move the needle to rarely, if at all.

Most auditors that I know are nice people. Most auditors that I know are pretty smart. Most auditors that I know are pretty honest. That said, most auditors that I know are being asked here to weigh in on matters in which they have no training and are frankly not likely to have it anytime soon.

What is the answer? I'm not sure. I'm not a fan of more regulation or more government control generally, but perhaps people who engage in review of compensation arrangements of public companies need some sort of licensure. Actuaries need it for many functions that we perform and we are also, as a group, pretty nice, smart, and honest.

I think that the PCAOB and SEC have used their unencumbered power to stretch too far. Perhaps, it's time to rein them in?

Public Plans, Beware What Your Actuary Can and Cannot Do

Thick laws create strange results. Consider for example the relatively new final SEC rule on registration of municipal advisers. As well as I can tell, the intent of this provision (Section 975) of the Dodd-Frank law was to ensure that those who are advising municipalities on things like bond offerings and investments. It never looked to me that actuaries were the target.

I don't know how it looked to the SEC. What I do know is that the final rule specifically exempted "retirement board" members from SEC registration. What I also know is that it did not exempt actuaries.

Why do I care and why might you care? It looks as if there are two situations where actuaries will be subject to registration as municipal advisers and if they are, they will be subject to all the attendant SEC rules:

• If your actuary performs an actuarial study in which the actuary sets any of the investment return assumptions or
• Makes any recommendation about how the governmental entity might address an unfunded liability including advice about the issuance of a municipal financial (debt) product,

 the actuary must register with the SEC.

It seems strange to me that municipal officials serving on retirement boards would be exempted, but actuaries may not be.

I guess it's not the first time that a government agency has confused me.

Pay Ratio Rule Explained ... In Plain Pithy English

Yesterday, the Securities and Exchange Commission (SEC) approved by a 3-2 vote along party lines a proposed rule implementing the pay ratio rule of Section 953(b) of the Dodd-Frank Wall Street Reform and Consumer Protection Act. Only here are you likely to see an explanation of the rule in words that you can understand and use. And, be forewarned, there is nothing in the rule that either serves to reform Wall Street or to protect consumers. In case you missed that, let me repeat in different words: the rule serves no useful purpose.

That said, I commend the commissioners and their staff for their efforts in crafting a rule that seems to well follow the statute while removing significant burden on companies in complying. I do not believe that the two dissenting votes object so much to the way the rule was crafted as they do that the rule was crafted at all.

Some of you will need some background on the rule. For those who don't, to refresh, 953(b) requires an issuer [of proxies] to disclose the ratio of the compensation as defined in Item 402(c)(2)(x) of Regulation S-K for the median-compensated employee in the company to that same definition of compensation for the CEO.

So, what does the proposal say? And, why, John, do you of all people, since you have clearly been anti-953(b) commend the commissioners and their staff?

The rule allows for significant simplification in the process. That said, multinational companies with decentralized payrolls may still spend millions of dollars complying with 953(b), but that still pales when compared with what they might have spent.

Disclosure Items

The rule requires the disclosure of three specific items:

1. Compensation for the median-compensated employee in the company
2. Compensation for the CEO (or PEO if you like the SEC vernacular better)
3. The ratio of 1 to 2, expressed as 1 to some integer, e.g., 1 to 377

Where a company uses sampling techniques, simplifying processes, or other estimation techniques (to be discussed later), the company is to describe these techniques, processes, assumptions, and methods in enough detail to be understandable to an investor, but not necessarily to satisfy an economist or statistician.

The issuer, at its discretion, may include other disclosures with it to assist an investor or potential investor in understanding the ratio. Such disclosures should be provided on a reasonably consistent basis. In other words, if in year 1, the issuer uses wiggle words to explain why the pay ratio is 1 to 999, then in year 2, the issuer should use similar verbiage to explain why the pay ratio is now 1 to 17.

Calculation of Compensation

As for other purposes in the proxy such as the Summary Compensation Table, compensation is to be calculated using the methods in 402(c)(2)(x). Without confusing the reader, this includes cash as well as the value of certain equity compensation and the increase in the present value of accumulated pension benefits, among other things. While the issuer (company) will already have calculated this for the CEO, the calculation for the median employee, hereafter referred to as Jane Doe, will also use the same methodology. So, to the extent that Jane is granted stock options and participates in one or more employer-sponsored pension plans (government-mandated plans are generally excludible), that must be included as part of compensation. 

Additionally, all disclosures are to be done in US dollars. Therefore, to the extent that (and I pity the poor company) Jane is paid in foreign currency, the value of such currency must be converted to US dollars. No adjustments may be made for cost-of-living differences in foreign geographies.

Who Gets Counted and How

All employees of the employer on the last day of the fiscal year are to be counted (leased employees and temporary employees of contractors may be excluded). Permanent full-time employees who worked less than the full fiscal year may have their compensation annualized. Part-time, seasonal, and temporary employees may not have their compensation annualized. And, as should now be clear unless you have slept through my first few paragraphs, foreign employees must be counted.

Simplifying the Process

This is where the SEC made us proud (hey, I was proud of them for this, but I guess I shouldn't put words in your mouth). Frequent readers will recall that I described 953(b) as a legislated disaster. (Actually, I need to thank Mary Hughes, my editor at Bloomberg/BNA for coming up with that description.) I am going to get a bit technical for a moment as I explain why to those people who have been hiding behind the rock in my blog.

Section 953(b) requires that companies disclose compensation for their median-compensated employee. Suppose a company has 999 employees. Then, the median-compensated employee (you remember her, Jane Doe) is the 500th highest-compensated. In order to determine who that is, the company would have to determine the compensation (402(c)(2)(x)) for each employee in the company. Currently, that is a by-hand process for the five employees usually disclosed in the Summary Compensation Table of the proxy. Doing it for another 1,000 or so would not be a worthwhile operation. Remember, for many companies, many employees will have many elements of compensation that take many hours to determine and that is way to many manys to be justified in the spirit of reforming Wall Street or protecting consumers.

The proposed rule allows companies to use sampling methods and and alternative forms of compensation in determining the median employee. For example, a company with only US employees, but having 250,000 of them might sample by selecting only those employees whose Social Security Numbers end in 22, 55, or 88. This seems unbiased and would still tend to produce a group of 7,500 people or so. I could go through the math for you (but, you're very welcome, I will refrain) to show you that the compensation of the median-compensated employee of the 7,500 will not vary significantly from that of Jane Doe. Further, in then determining the median of the 7,500, the company may (if it is appropriate for that company) use a convenient measure such as W-2 compensation specifically to determine which employee is the median-compensated one. Then, the company must determine 402(c)(2)(x) compensation for the CEO and that one other person.

Effective Date

I am used to reading IRS regulations. They usually tell you when you must comply with a regulation and they tell you using words (it may the only place they use such words) that you and I can understand. The SEC has chosen to do otherwise. In fact, after reading through their description three or four times, I am still not convinced that the actual rule says what their description of their own rule says. In other words, it is really confusing.

That said, I will use the SEC's own example. If the final rule were to become effective in 2014 and the issuing company had a fiscal year ending on December 31, then such issuer would need to comply for the 2015 fiscal year meaning that the pay ratio disclosure would need to be provide by 120 days after the end of 2015.

Request for Comments

Again, I praise the SEC. They actually want to get this right. They asked 69 questions requesting comments. While I don't entirely agree with their choice of questions, their willingness to ask them and to seek good answers is laudable. 

I will be commenting. I would encourage other interested parties to comment as well.

How Should Issuers Prepare?

The good news is that issuers have lots of time to prepare. It sounds to me as if you won't have to be disclosing pay ratios for another 30 months or so. That's a lot of time. It gives you time to get your ObamaCare ducks in a row before you think about this.

But, pay ratio disclosure will come. Here is a non-exhaustive list of things I think you should do.

• Figure out all the payrolls that you have company-wide
• Determine a measure of compensation that is (or something comparable is) reasonably and readily available such as W-2 or the foreign equivalent(s)
• Understand all the places and to whom equity compensation is issued
• Understand all the defined benefit (including cash balance) plans out there as well as nonqualified deferred compensation
• Prepare your foreign payroll administrators for the eventual need
• Develop a sampling method that is appropriate for your company (need help?)

You Have Questions?

The regulation and explanation combined are 162 pages. It's not exciting reading and in fact, does not have a good plot. I tried to present a readable and understandable summary here, but there's obviously much more. 

If you do have questions, you can ask me.

Was This Article Helpful or Useful or at Least Entertaining?

Please click the little +1 button at the bottom of the article and recommend it to your colleagues.

Pay Ratio Here, I Fear

The did it. By an overwhelming 3-2 vote, strictly along party lines (tsk, tsk), the Securities and Exchange Commission approved a proposed rule under Dodd-Frank Section 953(b). The two dissenters, Commissioners Gallagher and Piwowar, both issued scathing condemnations of both the statute and the proposed rule while the other commissioners praised it for the assistance that it will provide to investors and potential investors.

The rule has not been published yet, or at least, I can't find a copy, but I did take some good notes, so my readers get an early summary with surprisingly enough, some cynicism from your faithful blogger.

Issuers will need to provide three numbers:

1. Compensation (Rule 402(c)(2)(10)) for the median-paid employee
2. Compensation of the CEO
3. The ratio of 1 to 2

Despite this being what the law calls for, #3 is stupid. For those who are not sure what I just said, #3 is stupid. Suppose I told you that the ratio of compensation of the median employee to the CEO is .0073. What would that number mean to you unless you are mathematically facile? Is that a good number or a bad number? Should you be happy?

I repeat, #3 is stupid. The ratio should have CEO pay in the numerator and be an integer.

Haters of the statute did get a few breaks:

• Companies may determine the median employee via statistical sampling. After doing this, they must determine the 402(c)(2)(10) compensation of that median employee rigorously. Despite commentary from Commisioner Piwowar to the contrary, this is a huge concession to issuers.
• Compensation of permanent full-time employees may be annualized.

On the other side, there are these provision:

• Part-time, temporary, and seasonal employees pay may not be annualized.
• Compensation of global employees must be currency converted.

And, on the I'm not sure until I see the actual regulation side,

• There is no rule to say how statistical sampling must be done, but it must be reasonable and consistent.
• The dispersion of pay should be a significant contributing factor in determining the sample size for statistical sampling.

I understand that the novel which shall be heretofore known as the Dodd-Frank pay ratio rule is quite voluminous. Stay tuned here to get the most easily readable and entertaining reports on what it says.

Pay Ratio Rules are Coming

I've written about it many times. It lives in Section 953(b) of the Dodd-Frank Wall Street Reform and Consumer Protection Act. It's the pay ratio requirement. If this is new to you, you can get the background on it here. The New York Times wrote about it recently and finds no evidence as to how or why Senator Menendez (D-NJ) added that section into the Act. Perhaps this site maintained by AFL-CIO was a motivating factor.

I write about this again because of the rumor mill surrounding the SEC (I know, you can't find it in the supermarket tabloids).This rumor mill tells me that the SEC will provide its first guidance on this and other [Dodd-Frank] Title IX issues next month (September 2013).

Suppose you are an employer subject to Title IX (generally an SEC registrant), what should you do and what should you look out for?

If you are in the simplest of all situations-- that is, no employees outside of the US, no part-timers, no seasonal employees, no equity compensation, no defined benefit plans and no perquisites, your job is easy. CEO compensation and that of other employees is going to look a lot like W-2. You might even have a centralized payroll system on which you can run a report and easily determine who the median-paid employee in the company is. This could be a 30-minute exercise.

Oh, you're not in that simplest of all situations? It's going to be more complicated then. You will be faced with the exciting prospect of currency conversions, actuarial present value calculations, Black-Scholes calculations, gathering of data from multiple payroll feeds, and determination of the value of other benefits on which you usually don't place a price tag.

We think that one of the reasons that the SEC has taken three years to provide rules on much of Title IX is that they have not been able to figure out what makes sense. After having received thousands of comment letters, many suggesting simplifying procedures, apparently they are close. We will see soon what the rules look like.

In the meantime, if you fall into the group of companies that has one or more of the "problem areas", you will want to see what the rules say and see how they will affect your company. When the rules are published, you can read my analysis here. It may not cover every legal detail as the ones from law firms do, but I hope that it will give you practical advice.

And, now or then, if you'd like to get a head start on how to deal with this, e-mail me and we can discuss how to handle your situation.

It's not going to be pretty, I'm afraid.

Compensation Risk

I was reading Mike Melbinger's blog today about compensation risk assessments (if you want to read online legal analysis of compensation issues, I strongly recommend his blog) and I got to thinking that oftentimes, the people who may be assisting clients with this assessment may not know much about risk. You see, in evaluating compensation risk (and the SEC doesn't really tell us what that means), companies are to look at all elements of remuneration for both executives and for other employees. So, that includes things like deferred compensation which includes both qualified and nonqualified retirement plans.

I've written a lot about risk in retirement plans from the employer standpoint. How much cost variability is there? Does this benefit properly align with corporate goals? In an enterprise risk framework, where do these plans fit in? Is there compliance risk? Is there risk associated with having retirement benefits that are so large that you can't get employees to leave when you'd like them to? Is there risk associated with not having a defined benefit plan when some of your competitors for talent have them?

That last question is confusing, isn't it?. But, think about it. Defined benefit plans to favor older workers, not because they are discriminatory, but because of the shorter discount period until retirement date. So, if you sponsor a generous 401(k) plan and your top competitor sponsors a generous defined benefit plan, then a knowing employee might work for you until they get to be about 45 and then just when they really know the business, take off to your competitor who has a generous defined benefit program. It just makes sense.

So, compensation risk isn't only what you probably think it is. If you want to do a really thorough analysis and consider all of your rewards programs, consider people who have sufficient expertise to help you through the process. You might just learn something about your programs that you hadn't thought of before. And, it might be really useful.

More on CEO Pay -- Whose Fault Is It?

OK, I'll say it -- US Chief Executive Officers, as a group, get paid a lot of money. There! Are they worth what they get paid? Some are, some are not. Gee, you're not surprised yet by anything I've said?

My question to you or to anyone else is how did this happen. I heard somewhere this week that back in the 80s, US CEOs had pay roughly equal to 30 times the median pay for US workers. The same person said that multiple is now in the 300s or 400s. Assuming that you think that is too much on average (and I do think it is), is it my fault that it got that way? Is it yours?

Times have changed. Compensation packages have changed. Some executive compensation consultancies found for a while that the best way to get hired was to be able to show that they could get the executives the most money. There was no independence of compensation committees back then, at least it wasn't particularly required.

Again, back in the 80s, how were CEOs paid? For most, they got a base pay and they got a bonus. Some received equity awards or grants, but that was nowhere near as common.

Three things happened:

1. Companies started giving mega-grants.
2. The country at large and the politicians decided CEO and other executive compensation was too high.
3. Congress was smart enough or not to put Section 162(m) (the million pay cap for deductibility) into the Code.

My memory isn't perfect on this one, but this is my blog, so I can write as if it is. The first mega-grant that I remember was paid by The Coca-Cola Company. Mr. Goizueta became the CEO of Coca-Cola in 1980 at the age of 48 and he was a superstar. The Board thought so, the Compensation Committee thought so, shareholders thought so, Wall Street thought so, and competitors thought so. Coca-Cola wanted to lock Mr. Goizueta in and to do so, they gave him an award of 1,000,000 restricted shares. What this meant, roughly speaking, was that if he stuck around for 10 years, he would get one million shares of KO stock which traded around $83 per share at the time. 

That was a lot of money. Wall Street thought that Coca-Cola had made a great decision.  

So, other companies followed suit. Executive compensation consultancies and practices within larger consulting firms actively trained their people on the intricacies of mega-grants. 

Two things happened. The economy plummeted as it does from time to time, and as often happens when the economy plummets, people think that executives make too much money (oftentimes they do). So, Congress "solved" the problem. 

The Fools on the Hill (Capitol Hill) added Section 162(m) to the Code saying that pay for the top five executives in excess of $1 million in a year was generally not deductible on the corporate tax return. But, they put in an exception for certain performance based pay. 

Ah, a loophole! Who woulda thunk it?

In the Internal Revenue Code, where there's a loophole, there's a loop to put through it, and virtually everyone could find this loop. So, CEOs of large companies everywhere suddenly had base pay at or below $1 million and complex performance clauses that could pay them lots of money, especially when the stock market was heading up. 

Finally, the SEC entered the picture. As they refined and further refined the proxy rules for registrant companies, the SEC decided how pay should be defined. So, now, where CEO compensation can decrease significantly if interest rates are high and the stock market is low on the last day of the company's fiscal year, or conversely, compensation can increase because interest rates plummet on December 31 while the stock market hits new highs, different groups have the opportunity to cherry-pick their data.

In any event, according to the data I heard this week, CEO pay is now a multiple of 400 or so of the median worker pay. I don't know where the number came from. I'm not sure that I care because I know that the number is a contrived one. On the other hand, that multiple is too high by anyone's standards.

It's not my fault, though. Is it yours?

SEC Approves Final Say on Pay Rules ... Narrowly

Yesterday, by a 3-2 vote, the Securities and Exchange Commission (SEC) approved its new final rules implementing the Dodd-Frank 'Say on Pay' provisions. Before discussing the key provisions of the new rules, let me digress by saying that compared to other government agencies such as the IRS and DOL, the SEC has what I view as a strange process. Presumably with significant input from Commissioners, staffers write the rules which are then voted on (usually at an open meeting) by the five commissioners. At the open meeting, there is enough discussion that attendees (live or by some sort of conferencing medium) can get the gist of what is in the rules. Then, they publish the rules.

For those who don't normally play in this arena, 'say on pay' is the common term for allowing shareholders of an SEC registrant the right to, at least triennially, express their opinions (non-binding vote) via an up or down vote on the compensation of top executives.

In any case, here are the keys as I heard them:

• Proxy issuers must give shareholders all four available choices for a say-on-pay vote. That is, they must have a choice between annual, biennial, triennial, or abstention.
• Companies must explain in their Compensation Discussion and Analysis (CD&A) what consideration they gave to the shareholder's say on pay vote.
• Future shareholder proposals on say on pay or say on pay frequency may only be excluded if the company adopts the frequency indicated by the majority vote (NOTE: I'm not sure what happens if no choice gets a majority vote, it could be that there is only a plurality).
• The requirement to report the results of these shareholder votes is to be postponed to 150 days after the annual meeting, at which point it should be filed in Form 8-K.
• Small companies get a 2-year delay in effective date.

There is one other (in my view) strange thing about the way these SEC rules evolve. Periodically, the staffers will receive questions from registrants. They publish their answers on the SEC website, and once published, by some magical process, those answers appear to become part of the final rule.