09/09/2024
In August, proxy advisor CGI Glass Lewis released an article (see HERE) discussing the “nearly universal” imbalance in how Australian executives value their STI and LTI opportunities. In this article, we examine some of the factors contributing to this imbalance, leveraging insights from behavioural economics.
Summary
Glass Lewis’s proposition is that even for a comparable quantum of STI and LTI, executives tend to value STI more highly due to their greater immediacy, higher expected value, and lower associated risks, even when LTI might have a higher overall maximum opportunity at face value.
LTIs are deemed to be riskier for multiple reasons:
- They most commonly depend on a Relative TSR (RTSR) vesting condition (used by roughly 68% of the ASX 300), tied to unforeseeable market conditions and often non-vesting for sub-median performance;
- Their long-term nature increases their exposure to unexpected events and market fluctuations;
- Executives often feel they have less control over LTI outcomes, especially when measured against external factors like RTSR; and
- Unvested LTI faces the risk of forfeiture if executives leave the company (possibly lapsing in full, or on a pro-rata basis for “good leavers”).
Shareholder-imposed constraints reinforce some of these risks, such as the requirement that RTSR-linked awards not vest below median performance, with the aim of preventing unmerited payouts. This may inadvertently shift executives’ focus to STIs.
Behavioural economics offers potential ways to remediate these issues, through concepts such as hyperbolic discounting and the endowment effect (covered in detail later), but it has yet to be fully utilised in practice.
Issues with Relative TSR and Other LTI Metrics
RTSR usually compares a company’s TSR performance with that of peer companies (often a bespoke peer group of comparable companies, or an index tracking either the broader market or a specific industry) over the entire performance period (typically 3-4 years).
The following vesting scale is typically employed:
- No vesting for TSR performance below the median of the peer group;
- 50% vesting at median performance (treated as “threshold”);
- 100% vesting at upper quartile (75th percentile) performance, with straight-line vesting between median and upper quartile performance.
Assuming a neutral view of a company’s performance, there is an equal chance that the RTSR outcome will fall into each quartile, leading to an expected RTSR vesting outcome at 43.75% of the maximum opportunity, and a 50% chance of no vesting at all.
Not only is an executive’s expected LTI value far lower than the LTI’s face value; there is a large risk they will receive nothing. Even though most companies don’t align with the assumption of equiprobable outcomes in each quartile, shareholder requirements of non-vesting below the median cause these concerns to hold true for all but the best-performing companies.
Other common shareholder guidelines can further compound these poor prospects of vesting, such as:
- Positive absolute TSR gateways;
- Non-TSR metrics should be set at challenging targets at grant, with post-grant adjustments of the targets discouraged; and/or
- Equity awards should always be hurdled.
Forecast risk and uncertainty also pose problems for non-RTSR metrics, which may commonly include earnings and absolute TSR. The process of setting reasonable targets for these measures is complicated when they are being set multiple years ahead, due to increasing uncertainty over time, and a heightened risk of falling “below threshold” relative to STI metrics.
“Good Leaver” Treatment and Pro-Ration of LTI
Executives face the risk of their “on-foot” unvested LTIs being forfeited upon ceasing employment. Under “good leaver” circumstances, where executives leave in good graces (e.g., retirement), LTI awards may be pro-rated based on the portion of the performance period served. In other “bad leaver” cases, unvested LTIs may lapse in full, along with any deferred STI elements.
The scenarios above represent partial “pro-rata” and “full” forfeitures of unvested LTI. Alternatively, forfeitures could be avoided entirely with unvested LTI remaining on-foot to be tested in the normal course, even after executives depart. This “no-forfeiture” approach was encouraged by our article on good-leaver LTI treatments, published in May (see HERE). It outlined some guiding principles for determining when a good leaver’s unvested incentives should be pro-rated.
Behavioural Economics Considerations
The phenomenon of immediate awards being more “motivating” than delayed awards is referred to by behavioural economists as “hyperbolic discounting”, a form of time discounting. This drop in motivational effect over time is further magnified when the delayed awards are at-risk.
A 2020 behavioural economics research paper (see HERE) discussed these issues in depth, considering the effects of both time discounting and risk discounting on the perceived value of executive incentives.
The paper estimated the time discount rate for delayed awards to be 33% per annum. Using the hyperbolic discounting formula (see HERE) with this discount rate, the time discounting effect of delayed awards would be 1/(1+33%*D), where D is the delay period in years.
Under this model, even assuming guaranteed vesting, an award deferred for one year would have about 75% of the perceived value of the same award vested immediately; after three years (the typical vesting period for LTI), this drops to around 50%. An equity grant worth $2m vesting in 3 years would therefore have a similar perceived value to an immediate $1m grant.
Considering this time-discounting effect together with the riskier nature of LTI grants with performance-based vesting conditions, it is clear why executives frequently value STI substantially more than LTI.
The paper proposed to address these problems with LTI by ditching it altogether, remunerating executives solely in generous salaries and annual cash bonuses, with the requirement that bonuses be invested in company shares until sufficient “skin-in-the-game” is attained (i.e., until a minimum shareholding requirement is reached). This would provide an alternative way of aligning executives with shareholder interests to TSR-based vesting conditions.
One of our articles from 2016 (see HERE) looked at the problems that time discounting and risk exposure cause for incentives, focusing on issues posed by TSR-based metrics. The article scoped the possibility of paying LTIs in smaller, more frequent amounts, to enhance LTI’s motivating effect.
Another 2017 article (see HERE) considered the insights of Nobel laureate economist Richard Thaler, and how they could be potentially incorporated into executive remuneration practices.
One of Thaler’s insights most pertinent to executive pay is the “endowment effect”, also known as “divestiture aversion”. This refers to a cognitive bias whereby individuals tend to value an object that they already own more highly than they would have valued the same object had they not already possessed it.
The endowment effect corresponds to “loss aversion” associated with ownership. According to loss aversion, individuals tend to be more loss-averse than gain-seeking for an equivalent amount (the pain of losing $100 outweighs the pleasure of winning $100).
LTIs are an obvious area of potential application for these concepts. Consider the following two approaches to vesting the same LTI awards (with the same quantum and performance/service conditions):
- Vest the LTI awards immediately, and enforce trading restrictions upon them over some period, with the risk that some or all of the awards may lapse during this time if performance/service conditions are not met; or
- Grant the LTI awards, with only the mere promise that they will eventually vest, contingent on continued service and performance conditions.
According to the endowment effect, the first approach would be substantially more motivating than the second, despite the only difference in the approaches being a different framing of how the awards are vested. Unfortunately, the second approach reflects conventional practice, and behaviourally informed vesting methods like the first approach are seldom employed.
While these behavioural economics concepts have already been applied to malus provisions, they have yet to be fully utilised in executive pay structures.
© Guerdon Associates 2025