27/04/2012
Many ASX listed companies still allow for re-tests in their long-term incentive (LTI) plan. A re-test is a provision in incentive plans that allows performance against pre-defined targets to be assessed again at some time (or times) after the initial assessment.
Re-tests have been criticised because they allow executives to have two (or more) “bites at the cherry”. That is, a lucrative reward is made easier if executives are allowed to “sit the exam” on multiple occasions. The case for re-tests was not made easier by egregious practices. For example, there were instances of companies with a relative total shareholder return requirement that allowed performance to be tested daily if the initial test failed until, in effect, market vagaries allowed the incentive to vest. Fortunately, the extent of these egregious practices has virtually stopped.
But the question for board remuneration committees remains – should LTI plan re-tests be permitted? This article attempts to help answer that question, and concludes with a remuneration committee checklist.
External stakeholders
With the advent of the two strikes law most remuneration committees consider the views of external stakeholders. But, as on most other executive pay issues, there is no clear-cut answer. If the listed company has a significant proportion of UK and European institutional shareholders it is likely that a “re-test” policy will not receive their support. But this has broader implications. The two proxy advisory firms with a high proportion of UK and European clients will be more inclined to recommend “no” votes for policies with LTI re-tests. As this international market has become more important to ISS and CGI Glass Lewis, the weighting given to re-tests is likely to increase (ISS has indicated its new guidelines to be issued in June will take a harder line on this issue). While oriented to an international market, any “no” vote recommendations will also be distributed to domestic clients, and will no doubt have an impact.
Interestingly, proxy advisers to the purely domestic market (Ownership Matters and ACSI) have a more sanguine view of re-testing. Providing the extent of re-testing is reasonable, applies over an extended performance period and does not make the performance task easier in any material sense, it will likely receive their support.
While domestic institutional investors’ guidelines vary, most would consider re-testing on a case-by-case basis and, to an extent, they will be influenced by their proxy advisers. Nevertheless, if the rationale for use of re-tests is sound, and boards engage with domestic investors, reasonable re-tests will receive support. However, this will not be the case with UK and other European institutional investors.
External factors impacting results
A valid reason for re-tests was the impact of factors over which management does not have control. For example, an airline or refinery company’s results and TSR would be impacted by the volatile cost of oil. A mining company’s TSR would be impacted by the volatile commodity prices for the ore it specialises in. Companies with significant off shore operations will be buffeted by exchange rates.
But the counter argument is that a re-test just allows an opportunity to benefit from this volatility on the upside.
Companies that have TSR regularly impacted by volatile external factors do not need to introduce a re-test to more fairly compensate executives for factors beyond their control. One way to deal with this issue is to risk adjust the TSR for the company and its peers (see HERE)
Getting the most out of an expense
The accounting standards dictate that relative TSR LTI expensing will be incurred whether the reward vests or not. Faced with this, an argument for re-testing is that it extends the “motivational life” of a LTI, such that it will continue to have an impact on retention and motivation after the initial test. In effect, re-testing provides more “motivational” impact for the expense that has to be incurred. From this point of view, it is in shareholders’ interests.
Not two bites, but bobbing for fewer apples
A re-test usually retains the level of achievement required, but extends the performance period. For example 12% average compound EPS growth over three years is re-tested for 12% average compound EPS growth over 4 years. But if the executive team achieved just 10% in that 1st three years, they have to achieve 18% in the 4th year to achieve a satisfactory re-test result. This is not easier (i.e. the two bites analogy). It is less probable and therefore harder (i.e. analogous to bobbing for fewer apples). Hence, the perspective of many Australian institutional shareholders is that this is fair enough. In fact, it is difficult to understand UK and European investors’ inability to accept re-testing for non-market (i.e. non TSR) measures.
Do not tempt fate
Shareholder antennae are raised if certain re-test practices are employed, inviting a “no” vote. Foremost among these is continuous re-testing of market based measures such as relative TSR. If TSR, or relative TSR, is tested often enough, it is likely that general market and your company’s stock volatility will allow the incentive to vest.
Another practice likely to raise the ire of investors is the shortening of the performance period to the period between tests. Hence a LTI subject to a 3 year relative TSR test may have a re-test based on the intervening 6 months’ relative TSR performance, rather than 3.5 years.
A board remuneration committee re-test checklist
- Is the measure subject to volatility from external factors outside of management control? If yes, there is an argument to consider re-testing.
- Can the measure be risk adjusted to account for the uncontrollable factors? If no, or not easily, there is an argument for re-testing.
- Is the measure a “market based measure,” i.e. subject to share price, which under the accounting standards will require the company to record an expense for the incentive payment even if it does not vest? If yes, there is an argument for re-testing in order to extend its motivational life and get the most out the expense that is going to be incurred in any case.
- Is the re-test at least as difficult to achieve as the initial test? Assuming the performance period is extended, non-market measures, such as EPS growth, or return on capital employed, require an outcome in the additional time period far exceeding the outcome in prior periods in order to achieve the benchmark. That is, the task is harder. In contrast, most relative TSR measures tend not be harder to achieve with each successive re-test.
- Is there any evidence of egregious practice? These include continuous, or daily, re-testing of TSR results to ride the crests as well as the troughs of market volatility, or the shortening of the test period to the gap between tests, rather than testing for performance since the grant date. If so, these plans are unlikely to be acceptable to domestic or foreign shareholders.
- What proportion of shareholders is in Europe/UK? If a large proportion, you should expect at least a similar proportion to vote no on executive director grants and the remuneration report.
- Are you skilled and experienced in shareholder engagement? If you are, and the arguments for a re-test are valid, and you have a mainly domestic shareholder base, then you have a good probability of shareholder support.