Managing HR Risk: Micro Versus Macro

It's an interesting question, at least I think it is; when looking at risks inherent in benefits and compensation programs, should a company look at them on a macro basis or a micro basis? Or said differently, do lots of small unattended risks add up to a large risk that if it were in a single program would be viewed as untenable?

One could probably look at some of the variety of public entities which either are seeking or have sought protection under Chapter 9 of the United States Bankruptcy Code. Additionally, we could examine those that are probably considering such action whether or not deliberations have been in the public eye. What makes them especially relevant to this question is that in so many cases, the costs that are causing these cities and counties to go bankrupt are related to benefits and compensation. They are, after all, the largest budget items for many of these entities.

How did they get to this point? We could get into a political debate here, but that is not my objective. I know that some of my readers lean left while others lean right. We could get into a debate about public unions, but again that's political and one sentence later, it remains not my objective.

Many cities, counties and other government entities provide generous benefits. The often include defined benefit pension plans that would be unheard of in the private sector, health care benefits that would be considered "Cadillac Plans" under the Affordable Care Act (ObamaCare), large amounts of bankable paid time off sometimes as a single quantity and sometimes as vacation time and sick time separately.

The accounting profession through GASB standards has brought attention to the liabilities associated with these costs. Funding of these obligations, however, is usually subject to municipal or state law and there have been some notable cases where local government may have played fast and loose with those laws.

Why do I raise these issues? Benefits were not always as generous as they are today. But, they creep up. In good times, they are enhanced. Look, we have a surplus this year, we can afford to enhance this benefit. Each time this is done, there is a new, albeit small, risk added to the total risk pie.

Suppose a public entity has a $1 billion annual budget and this new enhancement only increases annual costs by $1 million. That's only 0.10% of the annual budget. Surely, it must be affordable. And, frankly, if you thought about it on a personal level, you would probably agree. If you have an annual income of $100,000, a $100 annual increase in costs would not change your life. It's only about $8.33 per month. It's less than 3 cents per day. But, you all know what happens, you don't just take on a single $100 increase, you take on a bunch of them and all of a sudden, you have a meaningful increase in costs and with it, a meaningful increase in risk. So go the public entities as well.

This sometimes happens in the private company world as well. Because of things like ERISA and the Internal Revenue Code, there are more funding rules and they cannot be ignored. But there are lots of benefits and compensation programs that need not be funded. You know what, they add up. Have too many of them and it starts to affect the bottom line. But when companies look at this on a micro basis, there doesn't appear to be a problem. You might hear a conversation that goes something like this:

"We have annual revenues of $1 billion, so where is the problem?" "I don't know, we haven't changed anything with meaningful cost." "Maybe our consultants messed up. When I look at our changes over the last few years, they didn't price any of them out to cost more than $500,000 per year [they forget to mention that there have been 25 such changes over the last 10 years each with estimated cost projections, but that all of those estimates come with inherent volatility and therefore risk], so maybe we should look to see where they messed up."

Suppose the company had looked at all 25 changes together. And, further, suppose they had looked at them stochastically and considered the dreaded left tail -- the worst 5% of all plausible occurrences. Wow, this entails some real risk!

So, while it's okay to consider changes that have costs and to be a bit more cavalier about ones that have small costs, companies should always consider them in a more macro context. What other changes have been made recently or are they considering? What is the interplay between the costs and attendant risks? How do they correlate? Will the costs escalate at the same time as business has a tendency to go bad?

Food for thought? What do you think?

What is the Risk-Free Rate?

For years, when looking at an expected rate of return on a pension trust, I have been asked to look at things like real rates of return, risk premiums, and the "risk-free rate." The risk-free rate has always been defined as the yield on US Treasury debt instruments, the safest investment in the world. Yesterday, the fine folks at Standard & Poor's gave a negative outlook on the US. They said that this means that there is at least a 1 in 3 chance that the United States' AAA credit rating will be downgraded within the next two years. S&P gave as its reason the potential inability of the government to get together and find a way to control the spiraling federal debt.

There remain a number of debt instruments, according to S&P, with a solid AAA rating. Interestingly, they all seem to yield more than US Treasuries. Should they become the risk-free rate? What is going on here? What is the risk-free rate?

I spoke with a few friends and acquaintances with more economic training than I. The consensus was that this was simply S&P's way of lighting a fire under Congress' and the President's collective posteriors. Surely, no US-based company can be more credit-worthy than the country in which it is domiciled. If the US economy is that weak, then how can individual companies be that strong?

As I meander back to my more customary topics, I look at the implications for pension plans. Surely, this action by S&P will cause the US borrowing rates to increase. This, in turn, would suggest that the yield on high-quality fixed income investments will increase. But, this will cause discount rates on public and corporate pension plan liabilities to increase which will decrease those liabilities and give the plans that support those liabilities better funded statuses.

What? Does that mean that this is a good thing? Are the collective state and local and pension plans really far less than $3 trillion underfunded?

It seems that this is a quandary worthy of the legendary Scotsman immortalized by the Bard of Avon: "Fair is foul, foul is fair."

In any event, this tells me that its time for the United States to employ some of the strategies that many have been preaching about at corporate levels for the last two decades. Our country is a large enterprise. Is it time for us to practice some sort of enterprise risk management? Should the Department of Homeland Security report up to the Secretary of Risk? Or is the Secretary of Risk just another name for the President?

I don't know, but perhaps it is useful food for thought.

On The Dynamics of Investment Committee Decisions

In a September 2010 survey conducted by Vanguard, more than 80% of investment committee members surveyed said that the knowledge level of their committee was above average and more than 60% said their committee seldom makes a mistake.

Wow!

I could spend hours now harping on the US educational system as the reason for this (grade inflation where a 'C' meaning average is only given to the worst performing students, but that's for another day and another blog). This is amazing, though ... or is it?

Let's consider why, and then we'll look at a potential solution. Who chooses committee members? The CFO? The Treasurer? Someone similarly situated? The Committee Chair? No matter which one of those it is, it's probably one of them. Who do they choose for the committee? In most cases, they probably choose people that they view as being a lot like them -- their proteges, for example. If that's the case, then a lot of the people on the committee will have gotten a lot of their knowledge from the same place.

Hmm! That's not good. That means that they may have similar biases. They may be serving on this committee chaired by their boss.

Hmm! That's not good. Through peer pressure, group dynamics, inertia, and many other dynamics, people tend to think that their group's ideas are the best, especially if their group builds consensus. Building consensus  generally is good, but what happens if the consensus is wrong? What happens if the consensus is under-(or un)-informed? What happens if the Committee Chair steers the committee toward his or her bias?

I know. This never happens on your committee, but you do know that it happens on most other committees. In fact, I've attended committee meetings where this happens. Sometimes, it produces good results, but all too frequently, it produces bad results based on what the committee thinks were outstanding decisions.

Let's put this in a defined benefit plan context. In order to properly invest defined benefit assets, the committee needs to understand both the assets and the liabilities. (Not doing so increases risk for the enterprise.) Yet, in my experience, very few committee members understand both sides of the equation (plan assets and plan liabilities). These people have other jobs. This is not their area of expertise. That being said, there are plenty of committees and committee members out there who just don't care. They have their committee comprised of some of the smartest people they know (including them, of course). Because they are smart, they will make the right decisions.

Consider an analogue. Suppose WeFlyHigh Airlines was considering which type of airplane to buy to add to their fleet. Should they buy the newest Boeing jumbo jet or the competing Airbus. Do you think they would have the Investment Committee members making that decision? Why not? They are smart people. Surely, they would make the right decision, wouldn't they? No, those decisions are usually made, or at least informed, by true experts. Shouldn't these committees also be informed by experts?

I promised you a solution, didn't I? And perhaps I've already shown my cards.

Hire an expert. Don't bring them on full time. But, this is why consultants exist, isn't it?

How should you choose your consultant? Look for these criteria:

• As part of their engagement, the consultant will educate the committee.
• The consultant understands the asset side and the liability side.
• As part of the liability side, the consultant understands the subtle changes to the liability profile that plan design changes and employee population shifts can cause.
• The consultant has a track record of making changes to the thinking of the committee.
• The consultant will challenge the decisions of the committee.
• The consultant can demonstrate a track record of giving different advice for different plans because those plans had different characteristics and were sponsored by different companies with different goals.
• The consultant freely admits that they are serving in the role of fiduciary.

I expect I'll write more on this topic. In the meantime, I'd love your comments.

Risk Management Isn't Just for Qualified Retirement Plans

Do you work for a company that has an active risk management policy? Do you consult with companies that manage risks or should? Are you in a benefit plan that may or may not manage risks? Then, perhaps this is for you.

Much has been made of risk management in recent years. During the economic downturn that started sometime during Bush (43)'s second term (I'm not going to argue about specifically when it started) and that is still continuing (or is not depending on which "expert" you believe), every company that I am aware of talked about risk management in earnest. Some actively did something about, some just talked about it, but it became a buzzword (I'm sorry, but a buzzword can be more than one word in my blog).

Let's focus on employee benefit programs, both broad-based and executive. Most everyone out there does some sort of risk management in most of their welfare benefit plans. Their health care plans are often fully insured, or if not, they at least have some sort of stop-loss insurance in place. And, they know that they can increase the employee portion of cost-sharing next year. With regard to other welfare benefits, LTD plans are often fully insured, life insurance plans as well. Think about them, in virtually all of these plans, employers are pooling their risks.

Suppose we turn to retirement plans. I am going to look at them in four baskets:

• qualified defined benefit
• qualified defined contribution
• nonqualified defined contribution
• nonqualified defined benefit

Qualified DB

What a trendy topic to write about: risk. The word has been out for nearly 25 years now. Get out of defined benefit plans. Diligent readers (I'm sure I have at least one) will recall that I wrote several weeks back that certain DB plans (specifically cash balance) managed appropriately are less risky than 401(k) plans from the plan sponsor standpoint. Since nobody commented on this, can I presume that everyone who read it agreed with me? I;m not that foolish, but ...

In any event, anyone who deals with qualified plans has heard about risk management, LDI, and lots of other trendy terms. When I got into this business, more large companies than not sponsored DB plans, and extremely few did anything to manage their inherent risks. Now, only those who think that they are omniscient with regard to both interest rate movement and equity and fixed income prices do nothing.

Qualified DC

In my experience, very few companies even evaluate their risks here, but most companies have them. Consider these:

• Suppose an employer provides a matching contribution in their 401(k) plan and all of their communications to employees are successful. Then, employees will contribute more and employers will be on the hook for more matching contributions. Isn't this a risk? I think it is. How many companies forecast this under any, let alone many scenarios? Shouldn't they?
• Many private or closely held companies sponsor ESOPs. When a private company sponsors an ESOP, isn't there really only one way to pay out plan participants when they terminate with a vested benefit? And, isn't that to repurchase the shares? I know that there are some companies out there that perform (or have done for them) an assessment of their repurchase liability. For the ones, who don't, in my opinion, they are just rolling the dice.
• I've seen a lot of companies scrap their defined benefit plans in favor of a profit sharing plan. In doing so, they make an implicit promise to their employees (not all companies do this, but there are enough that do), that they will contribute at least some minimum percentage of pay to the plan on behalf of each employee. Suppose business is bad. Suppose there are no profits to support these profit sharing contributions. That's pretty risky. The old way of sponsoring a profit sharing plan, basing contributions on and sharing profits, is probably more prudent. 

Nonqualified plans

Why don't employers (as a group) manage their risks in these plans? Are the obligations too small to worry about? Is it because they are just executive plans and since they may not get ERISA protection, they are not worthy of risk management? Is it because they don't know how? Is it because they have never thought about it?

Let's go back a few years, say to some point before 1986 (there were sweeping changes to tax laws including the treatment of certain life insurance products). Nonqualified plans were much smaller than they are now. But the promises made in many of them were just plain silly. 

I'm aware of one former Fortune 100 company (the company no longer exists due to acquisitions, but its particular identity is irrelevant) that promised a return in excess of 20% annually in its NQDC plan. They funded the plan using COLI, and they could point to broker illustrations that showed that all was taken care of. [pause for me to laugh out loud] I'm sure that there were other companies out there that did similar things. Remember that whoever it was that designed the plan (internally) was going to benefit from that large rate of return. If they were at the level that they were involved in the design, they were probably at least 40 years old at the time and they were smart enough to know that it doesn't take too many years at a guaranteed 20+% rate of return to build up a pretty good nest egg. And, they also knew if they thought about it that the risks that they created for the company wouldn't become really apparent until after they had become a wealthy retiree.

Why didn't this company manage this risk better? They were using the same mentality that many others have used in a retirement plan investment context -- that of total return. And, their assumptions were overly optimistic.

I'm going to make a bold statement (it's not really so bold, but teeing it up this way gets your attention better). Whether it be on a micro basis (at the plan level) or on a macro basis (at the enterprise level), it is critical that companies manage their nonqualified risks. This means that it is incumbent upon them to set aside assets to appropriately manage those risks (whatever that means to the particular company). While it may seem prudent to make the play that, on average, minimizes financial accounting costs, this is often wrong. While it may seem prudent to take the position that managing cash flow, in a way that on average, minimizes that cash flow, this is often wrong.

I'm going to attempt something drastic here. I'm going to try to insert a graphic in this blog (people over the age of 50 should generally not resort to such technological indulgences).

Tell me, in this matrix, which risk do you really want to take additional steps to actively manage? If I ask people (limited to ones that I consider to pretty intelligent), they assume that this is a trick question. Of course, they want to focus on the northeast corner -- the one with high risk and high likelihood. Think about it. Aren't these the risks that they are already very actively managing? With regard to these risks, companies tend to be fully insured, fully hedged, or at least fully something. They don't let these risks go unwatched. They know that they could bring down the company.

Does anyone remember what happened to BP in the Gulf of Mexico a few months ago? The likelihood of that event was small, but the downside risk was immense. Similarly, does anyone remember the performance of assets and liabilities together during the 10-year period that just recently ended? We had about 4 years of "left  tail events during that 10 year period (the left tail in a normal distribution is where you usually find the low probability, but very poor outcome events). In other words, 40% of the time, we had an outcome that models said would occur less than 5% of the time. What went wrong?

I could go on and on about what went wrong in terms of bad models, bad laws, bad accounting rules and the like, but that's not the point. The point is that not enough companies prepared for this low-probability event. Now, risk management in pension plans is all the rage (at least for companies that still sponsor pension plans). As time goes by and the rich get richer (they do, don't they?) nonqualified liabilities grow rapidly. And, as those liabilities grow, companies are actively managing the risks attendant to those plans, right?

WRONG!

Some have looked at this carefully, but you can probably count that list of some pretty darn quickly. Am I suggesting that companies formally fund their nonqualified obligations in a secular trust? Probably not, the tax rules usually don't work. Am I suggesting that they fund in a rabbi trust? Maybe, perhaps more than maybe. Am I suggesting that they somehow evaluate their low probability, high magnitude risks in their nonqualified plans and then quickly take reasonable steps to ensure that those risks will not get in the way of the successful operation of their company?

With due credit to Rowan and Martin's Laugh-In, you bet your bippy I am.

Do it now, and do it right, and if you're not sure how, let me help you do it.

Looking at the Bigger Picture -- Forecasting on an Enterprise-Wide Basis

I've been a consulting actuary for more than 25 years now. As I think back at some of the work that I have done and supervised, I realize how flawed some of it has been -- not the work itself, but the nature of the work that we were asked to do. We've been asked to do work in a vacuum, but independent of other parts of our client, its benefits program in total, or its human resources program in total. So, while we may have done an excellent job, the assignment may not have had the value to our client that it could have had we been engaged "properly."

I'm not saying that this was necessarily anyone's fault. Perhaps we were the retirement consultants and weren't being engaged for any other HR work. Perhaps the individual engaging us didn't have purview beyond the retirement plans. In any event, however, I'm taking a guess that I've had clients who looked at bad-case or worst-case scenarios that I presented to them and combined them with other bad-case or worst case scenarios in doing planning.

To illustrate, let me use a (hopefully) absurd hypothetical. Suppose the worst-case for a company's energy division occurs when temperatures are moderate, but the worst case for their agriculture division is when temperatures are extreme. Conversely, the best case for the energy division is extreme temperatures and the best case for the agriculture division is moderate temperatures. So, in total, the enterprise has hedged its risks well.

However, in our hypothetical, each division is left to its own devices to manage its own risks. So, the energy division spends money to "insure" against moderate temperatures and the agriculture division spends money to  "insure" against extreme temperatures. Each division has spent money to manage risks that were already managed on an enterprise-wide basis. Of course, this never happens in real-life, does it? WRONG! It happens all the time. I've seen it.

It's often difficult for all but the top executives in an organization to evaluate risks and "insurance" needs on an enterprise-wide basis. On the other hand, within the human resources function, for example, isn't the entire department its own sub-enterprise? So, as an example, might it not be a good idea to look at the effect of a 1% decrease in underlying interest rates on the entire HR function rather than just on the pension plan? Perhaps in a lot of those scenarios, as interest rates fall, so does inflation and wage increases. So, while pension plan costs might be increasing, compensation costs might be decreasing. In total, HR may be doing a fine job of managing its costs, but if the pension manager is asked to manage pension costs and the compensation manager separately is asked to manage compensation costs, each could be spending resources managing a risk that within the total HR department is already appropriately managed.

I'll comment more on this in the future, but for the time being, I just wanted to give some food for thought. For mid-level managers, the corollary to this is that the better you can view things on an enterprise-wide basis, the faster you will escape that mid level and get to a high level.

Think about it ...