Are CEO Compensation Markets Efficient?

Having been schooled at Chicago, considered an institutional icon in espousing free market and efficient market theories, I am always initially skeptical of things that fly in the face of efficient markets. This post by Douglas Smith, however, is excellent and reflects upon a WSJ journal article in which Treasury Secretary John Snow maintains the widening gap between high-paid and low-paid Americans reflects a labor market efficiently rewarding more-productive people.

Of course, I submit (perhaps a little unfairly given recent circumstances) Exhibit 5, to Douglas Smith’s list of exhibits. Joe Nacchio, former CEO of Qwest, was once in the top 10 of all compensated CEOs and earning in excess of $100 million/year. As a former shareholder in the firm, I think I saw my shares slip in value close to 50 times during his term (here’s an article from 2002 that adds some color to the craziness of the compensation situation). Although I recall Nacchio (or folks on his behalf or in defense of his raises) claiming that Nacchio deserved compensation competitive with what he could get elsewhere in the market, all I can say is that Nacchio was neither Larry Ellison nor Michael Dell in terms of delivering value to the shareholders.

This is not to say that I don’t believe in higher pay for senior executives. Executives have a lot of leverage created in part by their span of control. Tremendous value in excess of market comps can be created or destroyed in fell swoops. But questions remain as to whether businesses (and the people that run them) have put in place the right control and reward structures, as Douglas Smith’s post points out.

The Pricing Death Spiral

Today I wanted to write about perhaps one of the most important lessons I learned in managerial accounting (as opposed to financial accounting), and it is something that I have seen improperly applied in many organizations (both profit and non-profit). It can be a very subtle concept related to accounting and pricing, but it can have very bad effects. The situation is called the "pricing death spiral".

Imagine a company that manufactures and sells gumballs. Suppose that the gumball company produces and sells 100,000 gumballs a year. The cost of manufacturing each gumball is $0.01 per gumball and the sales and marketing costs are $1,000 per year. No other costs apply. If one tries to unitize sales and marketing costs, over the number of gumballs produced, the sales and marketing costs are an additional $0.01 per gumball. If the company sells the gumballs at $0.02 per gumball, it makes no profit.

Now suppose that that sales are falling and so the company only produces and sells 50,000 gumballs in a year with the same sales and marketing costs. Now sales and marketing costs are double when one looks at the per gumball price. Costs are then $0.01 per gumball (manufacturing) and $0.02 per gumball (sales and marketing). So the company needs to raise prices to $0.03 per gumball just to breakeven. Herein lies the death spiral of pricing. Raising prices in this context artificially creates less demand for products, and thus, fewer sales. Weak sales seem to generate higher costs which causes higher prices which creates less demand and even weaker sales, etc. If the company (incorrectly) tries to unitize the sales and marketing costs across the gumballs produced and then folds that into the pricing of its products, it can wind up in a self-reinforcing situation where it is raising its prices despite falling demand and without regard to the competitive environment.

Some variations of death spirals I have seen in organizations I’ve worked with:

  • a non-profit child care center has low enrollment (below breakeven) and raises monthly fees for next year’s class to amortize the administrative office costs across all children
  • a professional services firms prices its services too high because it spreads out the costs of G&A of all areas of the business into the cost structure of a services group of a smaller division of the company (which in turns creates less demand for services and fewer subcontractor partnerships than optimal)
  • the standard costing for a construction job is based on amortizing the entire cost of the field force the number of jobs performed – in a weak construction season, the next year’s costs are calculated to be too high to feed the pricing equation.

While the realities of an organization’s financials may dictate how closely cost structure should affect pricing, the error in the above pricing methods come down to this: the organization needs to separate out fixed costs (or sustaining costs) from variable costs. The organization also needs treat excess production capacity (or in the opposite case production levels in excess of normal operating levels) separate from unit costs.

Thus, in the example I had above on the child care center, rather than calculating the cost of providing care for a child by spreading out all G&A costs over the number of children enrolled, a unit cost per child should be calculated by spreading out a proportion of G&A by the child care center’s normal operating capacity. When the child care center runs below normal operating capacity, it runs a temporary loss equal to the costs times the number of children below normal operating capacity. The shortage in income could be made up by assessing the families of the children by a monthly (or lump sum) fee. One key benefit of pursuing this latter pricing mechanism is that the price of low sales is separated from the cost of production. Families of the children are only assessed fees to cover temporary shortages in enrollment, and the situation does not evolve into a case where the "permanent cost structure" is raised (thus forcing ongoing prices to be raised excessively).

The other cases of pricing errors I describe above are a little more subtle. In the second case, the error is that the company should again calculate variable unit costs for producing individual business unit services. While spreading out G&A for the business unit itself may be less suspect, spreading G&A from the entire corporation may be excessive (although computationally convenient) and should be viewed as a sustaining cost for the business. In the third case, the construction company would be better to allocate a portion of G&A to the individual job cost structure presuming that the construction company is operating at healthy levels.

So while there is no single, silver bullet for avoiding a pricing death spiral, first steps are recognizing the situation and then separating out the considerations so that more accurate responses can be taken (i.e., if production problem then solve production problem, if sales problem then solve sales problem -> improper accounting can mask the problem at hand). If one finds that the price of one’s products or services have more to do with the size of the finance organization in your company that the manufacturing or services costs, this may be a good warning sign.

The Nickel Tour

With all the stuff going around about the bird flu, I am reminded of one of the less glamourous management consulting projects I heard about (in general terms) last year in the turkey business.

Now in many operations projects, key goals are to improve business processes in dimensions such as:

  • average throughput
  • inventory backlog
  • peak capacity
  • quality
  • cycle-time
  • cost
  • risk/failure points

A company often has tons of business processes in place. Sometimes there may be a manageable set of predominant process flows, but then there can be a zillion microflows. One way for a consultant to get grounded in a situation in the face of this complexity is to go on a "nickel tour" with the client .

In the case of the management consultant I met with, the goal of current project was to reduce the number of injuries in the processing plants of one of the big turkey producers (I presume to reduce lawsuits, etc.). The automated equipment in certain sectors of the meat business, as I understand things, can be quite scary. Not for the faint-hearted for sure, some of the equipment used can separate the meat from bone (of entire animals) in matters of a few seconds. Imagine what can happen if your arm gets caught in the machine …

So day 1 the consultant arrives on the scene, and one of the plant workers hands the consultant a pair of rubber boots to go on a "nickel tour" of the plant. I don’t think the tour was of the slaughterhouse, but one can imagine that the scene was not everything a recent MBA grad dreams of doing as a consultant.

To generalize, in many nickel tours, the client walks the consultant through the backoffice, introduces sales personnel, has them sit in with customer service representatives, attend working meetings related to information technology user sessions, etc. The purpose is to give the consultant a ground floor view of what happens in the business (plus an opportunity to ask questions). The nickel tour helps to compress a complex view of the business into one short experience. While a lot of the tour can turn out to be a bunch of chit-chat and small talk, I have often turned the nickel tour into a very useful experience. The nickel tour can be a very valuable source for initial checkpoint information for the consultant (e.g., if the consultant sees large piles of inventory, frazzled or distressed workers, disorganized workspaces). A consultant may also meet people on the tour that can serve as useful sources of information later in an engagement.

On the flip-side, a consultant needs to be wary of "stage plays". This is a case where the nickel tour is not a real tour of operations, but a case where someone (within the client operations) has dressed up the situation to be different or better than it really is on a day-to-day basis.

In any case, make sure to think about giving or getting a nickel tour in a consulting relationship. Although it is not always possible or desirable in some cases to give a nickel tour, a nickel tour can really help consultants get a "live" feel for the business at hand.

Somewhere Between Goal Management, Client Facilitation, and CYA

As I work through my own professional goals plan for the year, I am reminded of an early lesson in management consulting. This technique has do more with firms that are implementation-oriented as opposed to strategy-based, and it is related to getting a client organization to move. I don’t know if there is a name for the technique, but for the purposes here, I’ll call it "progress to goal management".

The essence of the technique begins with the consultant working with the client organization to establish one or more measureable goals and then reporting on actual performance regularly with a gap diagnostic. Let’s say a goal is to create $5.5 million in annual revenue for a new start-up initiative (say $500K in Q1, $1M in Q2, $2M in Q3, and $2M in Q4). The consultant then works with the client to put a regular measurement system in place, say monthly. Suppose that by the end of Q1 that the client has only achieved 15% of the goal. The consultant should be working with the manager owning the revenue to report not only the numeric gap in performance but also a diagnostic of why things are off track (from both quantitative and qualitative perspectives).

Although the technique may seem obvious, in implementation consulting, one is often trying to diagnose problems or set up operations for things that happen below a corporate-level or business unit-level. The devil may be in the details and low-levels of the company so-to-speak. Thus, things like board-level measurements and control may neither be enforced nor visible. Other situations where measurements may not be readily available include setting up new business structures (e.g., new business line) or bypassing old business structures (e.g., where old methods too cumbersome or bureaucratic).

Other than trying to help the client (which is the primary goal, of course), the flip side of this is that the consultant is pulling a "CYA" in some sense. One can’t start a project and then six months later show up and simply report to executive management that the goal wasn’t met by the business unit or functional area management. Progress to goal (and gap) reporting is needed every step of the way along with mid-course control and corrections. In this way, the consultant separates the aspects of proper management control from management’s ability to execute.