Psych Safety, by Iterum

Lost in Translation: The Effects of Incentive Compensation on Strategy Surrogation

Choi, Hecht & Tayler (2012) · One paid measure, more surrogation

Choi, Hecht & Tayler · Primary Research · Measurement & Method, Critique & Boundary · The Accounting Review · 2012 · Open access

Open this neighbourhood on the live map →

Choi, Hecht and Tayler name a hazard of strategic performance measures: managers may 'act as though the measures were the construct of interest', which they label surrogation, after Ijiri, and test whether tying pay to measures changes it. Drawing on Kahneman and Frederick's attribute substitution, they argue that paying on one measure of a strategic construct makes that measure more accessible as a substitute, raising surrogation against a flat wage, and posit that paying on several prompts comparisons that highlight 'the limited ability of any one measure', lowering it.

In the main experiment 79 graduate business students designed a creature in the game Spore to a stated strategy: socialise with other life forms to become the dominant species. Four measures (sing, dance, charm, pose) updated as they built; each, they were told, mattered equally. One group got a flat $10, one $2 per level of sing, one $1 per level of sing, dance and charm. Then, unpaid, they chose from each of 18 pairs of modification packages the one they believed would best implement the strategy; the surrogation score is the share of relevant choices favouring the paid measure or measures, scored on the same measures for the flat wage group. In some pairs that choice gave up a higher total across all four.

Pay on one measure raised the mean score from 0.39 to 0.59 (one-tailed p < 0.01), and the effect held where the choice carried that cost. Pay on three measures gave 0.46, lower than one (one-tailed p = 0.02); on the costly pairs alone the difference was not significant, though it was marginally significant (one-tailed p = 0.09) when only pairs scored for both groups were compared. On the overall score three measures did not differ significantly from the flat wage, though on the costly pairs they were marginally higher (0.29 against 0.16, two-tailed p = 0.10). A follow-up of 50 participants, now paid for their choices on their ability to implement the strategy, again found less surrogation under three measures than one (0.46 against 0.61, one-tailed p = 0.01). Since the main experiment's choices were unpaid, they argue the single-measure result 'cannot be explained by a simple propensity toward wealth-maximization'.

They name the limits: a non-business setting, where 'it may be easier for managers to keep strategy in mind in more familiar environments'; the number of measures at which surrogation begins to ease is left open; and attention was self-reported, 'potentially very noisy'.

The paper does not name psychological safety; the bridge is ours. We read it as a caution for any psychological safety score tied to pay or reward: the risk it isolates is not a simple response to incentives but treating the number as the thing it stands for, which in their laboratory one paid measure invited more than three did.

Choi, J., Hecht, G.W. and Tayler, W.B. (2012) 'Lost in translation: the effects of incentive compensation on strategy surrogation', The Accounting Review, 87(4), pp. 1135–1163.

Save this paper to a bibliography, keep notes on it, and export citations. Create a free account →

Related psychsafety.com articles (5)

Related papers (2)

View the source paper →