Cost ≠ Value. Issues with the total compensation approach

In several sectors compensation professionals and the HR teams they serve talk about total compensation. When communicating compensation messages with employees and prospective employees they sum the cost of each component and present this as the value of the total package.

Unfortunately what is important to employees is not the fiscal cost of the package, but the perceived value, and they are rarely the same thing. The issue is that given the money employees would often prefer to allocate it in a way that gives them more perceived value. Economists describe this as a deadweight loss: a waste of resources that could be averted without making anyone worse off. Think about it another way, if instead of providing non-cash benefits a company paid just cash then a benefit package would only be efficient if the employee would choose to spend the money in exactly the same way as the employer had.

In a famous 2001 paper Joel Waldfogel attempted to calculate the deadweight loss that giving presents at Christmas generated over the alternative of giving the recipient the equivalent cash. He estimated that the deadweight loss was between 10% and a third. It is highly likely your total compensation is being similarly devalued by employees.

Why non-cash benefits can make sense

So the economic theory suggests that the most efficient way – that that creates the highest perceived value to employees – of providing compensation is just to provide the cash. However this misses part of the picture. Companies can often purchase goods and services at considerably reduced prices than employees can. If we take Waldfogel’s 10–33% estimate then if the company can purchase the goods on behalf of the employee at this sort of discount then the package could be more valuable to the employee than just giving a bigger salary.

How to measure what is a valued package

Techniques exist for measuring how various components of a package are perceived but they’re rarely used by employers to measure perceived value of benefits packages. They should be.

The class of problem that you need to understand is termed discrete choice: that is you have a maximum total resource to allocate and having more of one thing means less of something else. The only way of measuring preferences is to replicate this trade-off.

The problem of maximising the perceived value of a package of goods is one faced by marketers daily. The established technique for measuring this is a technique called conjoint analysis. It can be very effectively applied to the compensation problem.

The importance of choice

The most effective way of maximising the perceived value of a compensation package is to provide a wide menu of options and let the employee choose which they want. This menu should be extensive and it should be possible to change items on a reasonably regular basis as employees preferences change, and as they do their ideal basket will change.

The issue with this approach is that it’s expensive. The companies it arguably works best for are those with a large number of relatively similar employees who have a similar set of preferences. It’s used quite frequently in large professional services firms for example.

If you don’t want to do this the second best approach is provide a series of menu options which bundle benefits in a way that appeals to various segments within your workforce. The great thing about discrete choice experiments such as conjoint analysis is that the results they produce are differentiated – that is that cluster analysis is possible using the data.

Communicating benefits to employees

The issue of the difference between cost and perceived value of benefits makes communicating benefit value difficult. If you communicate a cost to employees but it is not something that they would have chosen then there is a real danger that the employee will view it as wasted money. One way of reducing this is to report not only your cost but also market cost – so that the employee sees that you have got them a bargain. Whichever way you communicate it is always worth doing material tests to understand the reactions to the wording. If you don’t want to do this directly with employees then it’s relatively straightforward to recruit test participants who are similar to your employees.

Negotiating a salary in a total comp world.

If you are a prospective employee and want to negotiate a good package then it’s always worth starting with the reported total compensation value of your current package. It is highly likely that the value to you will not be as high as your current employer is reporting due to deadweight loss. By starting at this level you can then negotiate a package that more effectively matches your perceived value and capture some of the deadweight loss.