To Benchmark or Not to Benchmark...

2009 will probably have been the most challenging year many HR people have experienced. Shrinking budgets, reduced resources and managing testing HR
agendas have taken their toll. And at the risk of being accused of being pessimistic, we predict that 2010 will be equally challenging for the HR community.

In a tight labour market, against the background of a recession, it is easy to think “Why do we need to benchmark pay? Labour turnover is low, pay settlements are low or non-existent and we are not recruiting.” However, there are a number of very good reasons why you should keep up-to-date with what is happening to reward practices in the market; a view which is supported by around 75 per cent of Paydata customers¹.

We believe there are six key reasons why employers need to keep an eye on what is happening:

  • Impact varies by industry;
  • Impact varies by region;
  • Changes in headcount impact survey data;
  • Bonus schemes and long-term incentives are under pressure;
  • There will always be a market for top talent; and
  • Taxation is likely to play an increasingly important role.

Impact varies by industry

Once you delve below the headline numbers of pay settlements, redundancy rates and so on, you get a better picture of what is happening in the labour market. For example, while the headline UK redundancy rate in Q3 2009 is 8.2 per cent, in manufacturing it is 15.5 per cent (from a high of 21 per cent) and in construction 26 per cent (from a high of 33 per cent). Conversely, services overall sit at 6.1 per cent and the public sector at just two per cent.

The impact of this on benchmarking data will vary depending on the type of survey. For a club survey comprising companies in the same industry sector or sub-sector the impact is probably negligible. However, for surveys using cross-industry samples, for example those looking at functional roles (HR, Finance, etc.) or those that have traditionally had large and diversified participant lists, the impact could be significant.

We suspect that most people using a survey flick past the introduction and go straight to the pay numbers. Our advice is you take the time to have a look at the analysis that almost all survey publishers provide about the participants and take a view on the impact this may have had on the reported pay rates.

Impact varies by region

The regional impact and the industry impact are in some ways inter-related. The fact that the West Midlands and the North East have the highest redundancy rates (at their peak 16.6 per cent and 16.9 per cent respectively) is almost certainly linked to the size of their manufacturing base.

It is very likely that changes in the regional mix of participants and employees will impact pay rates reported in salary surveys. This is probably true for all types of surveys including some of the industry club surveys that we produce. This is because all things being equal it is statistically more likely that the number of employees included from lower paying regions will have decreased as a proportion of the total sample. The impact could be that pay rates will appear to have increased at a national level even if the majority of participants have implemented pay freezes.

Changes in headcount impact survey data

The easiest way to explain this is to use an example. Let’s say in 2008 you had twenty jobholders in a particular role. By the end of 2009 you have made half of these people redundant. The ten people that remain in employment will therefore form your submission to the next salary survey:

  • If you have released your lowest paid employees (perhaps because they are the least experienced) then your average salary will go up; and
  • If you have released your highest paid employees then your average salary will go down.

So without making any changes to actual pay rates you have affected how you compare to the market. But what actions have the other participants taken? The market rates could also go up or down depending on how headcount has changed in each participating organisation.

The net result is that your position against the market might move one way or the other depending on what you and the other participants have done in terms of restructuring your headcount.

Bonus schemes and long-term incentives are under pressure

We know from the results of the Paydata UK Reward Survey that the focus on variable pay within the reward mix is increasing (45 per cent of participants). We also know that the current economic conditions mean that bonus payments made based on 2009 performance are likely to be smaller than in previous years (58 per cent).

The problem is there is a lag between what companies pay and what surveys report. This is because the surveys reflect the last bonus paid (which could be up to one year earlier):

  • When bonus payments are falling surveys tend to overstate variable cash; and
  • When bonus payments are increasing surveys tend to understate variable cash.

For this reason it is unwise to rely purely on survey reports without applying some judgement regarding the context.

It is also important to understand how the survey numbers work. A common mistake is to subtract the median base salary from the median total cash number and conclude that the difference is the average bonus. This is incorrect because in most surveys the two numbers are calculated independently. So the employee who is paid the median base salary is not necessarily the same as the employee who is paid the median total cash. The most useful survey data is either the actual bonus payments or actual bonus payments as a percentage of the on-target position in the bonus scheme.

There will always be a market for top talent

For a business to succeed in a tough market it requires a lot more expertise than when market conditions are more favourable. The best businesses also look to take advantage of the weaknesses of their competitors to gain long-term advantage. The net result means that even in a weak economy there is a robust market for top talent.

It is perhaps for this reason that the Paydata UK Reward Survey has consistently found that companies are looking to direct their pay awards towards their top performing people (83 per cent of participants). A caveat to bear in mind here is that these pay actions should be directed towards people whose underlying value to the organisation is growing fastest only when it can be shown that they are relatively underpaid in relation to their level of contribution. Part of this is about the internal labour market and part of this is about understanding what competitors pay for similar roles in the external market.

This is where benchmarking can play a central role in ensuring that the pay review budget is used to greatest effect. Cash is almost always the most valued business resource, and particularly so when times are tough. Benchmarking can help ensure you spend your cash as effectively as possible.

Taxation is likely to play an increasingly important role

Marginal tax rates in the UK have been relatively low for the last 20 years. For those with long memories the top UK rate of tax on employment income hit a record high of 83 per cent in 1974 which stayed in place until the tax year 1978-79. The final reduction to the current top rate of 40 per cent took place in 1988.

The net impact is that UK reward professionals have not overly-focussed on the tax-effectiveness of reward arrangements for two decades. However, this changed significantly when it was recently announced that the top rate of tax will increase to 50 per cent in 2010.

Governments around the world will no doubt look to increase the tax take (the proportion of GDP levied in taxation) in order to help reduce their debts. A good
example of action already taken is Ireland where the Government has increased all income tax rates.

As a result it is inevitable that tax-effective reward will become an important part of the agenda for the foreseeable future; and this is not just for the highest paid. There are already signs the Inland Revenue will look to limit the scope and effectiveness of “salary sacrifice” arrangements which are designed to reduce tax and/or national insurance costs for employees.

Salary surveys can help you keep up-to-date with the latest trends. The most important information is likely to be in the benefits commentary rather than in the
statistical analysis of the data.

Our recommendations

Make sure you understand the data
Review the survey participant lists and/or the analysis of participants by sector or region, compare to the previous year and take a view on what the impact is likely to be. Add in your own knowledge of what has happened in your industry or local market.

Review your data sources
Consider adding in new survey sources. Or if benchmarking against a specific peer group consider adding more organisations to your comparators, perhaps from related or complementary industry sectors.

Consider using upper and lower ranges instead of quartiles
Due to the nature of the statistics the upper and lower quartiles are more susceptible to changes in the survey samples and the market than the median.

One way to work around this is to apply a percentage adjustment to the median to create a target range. The size of this range tends to vary by job size. We have been operating our PAYanalyst database and our pay modelling approach in this way for a number of years because it provides more reliable results. This way of using the market data supports a move towards creating market competitive zones as opposed to overly-focussing on a particular percentile/spot rate. This approach is more likely to produce sustainable policy positions as market conditions change.

Focus reward spend in order to deliver the best return on investment
Employers need to model pay arrangements to get a better understanding of how individual contribution or value maps to relative pay. This means using benchmarking data and internal pay relativities in an informed way to maximise the return on payroll spend.

In conclusion

2009 was undoubtedly a tough year for business and especially for people in the HR community. But we believe that for many businesses it will come to signify the start of a change in the way performance is managed and people are paid. Market benchmarking can play an important role in this process as it provides part of the context for decision-making regarding what is appropriate for each business based on its circumstances.

¹Average response to Paydata UK Reward Survey during 2009 (conducted in January, June and September).