The role of performance feedback consistency in aspiration level adaptation: Evidence from a European grocery retailer

Abstract

This study investigates whether and how the impact of drivers of aspiration levels changes across the cases of consistent and inconsistent performance feedback within the context of a retailer. Analysis of internal corporate data shows that while past aspiration level and performance–aspiration gap positively influence the current aspiration level in the case of inconsistent feedback, performance feedback consistency changes only the impact of performance relative to peers. This study replicates past research in a different industry and country due to limited empirical evidence, introduces real-world complexity into aspiration theory, pinpoints performance–aspiration gap as the primary performance feedback, introduces a new sign for the impact of performance relative to peers, and reconciles its previously detected mixed impact. The findings suggest that organizational attention has an inward focus in the case of inconsistent feedback. The results also point out that leaders can trigger change through a performance outcome that lags behind the corresponding aspiration level rather than the performance of peers and eventually move their organizations toward high performance targets by starting with feasible rather than stretch goals.

Keywords

aspiration levels behavioral strategy change goals panel data performance feedback

Introduction

“The level of an aspiration is the target performance” (Shinkle, 2012: 422) and thereby the benchmark for success. It is “the smallest outcome that would be deemed satisfactory by the decision maker” (Schneider, 1992: 1053) such as 10% return on revenue. Aspiration levels guide firm strategy and conduct through their effects on compensation, control, and resource allocation. Thus, understanding and shaping the drivers of aspiration levels means understanding and shaping firm strategy. However, the research on the drivers of aspiration levels has been sparse and simplistic. There are only handful of studies (Lant, 1992; Mezias et al., 2002; Washburn and Bromiley, 2012) and these studies overlook real-world managerial complexities such as the role of inconsistent performance feedback in the adjustment of aspiration levels.

Managers frequently grapple with inconsistent performance feedback that is simultaneously positive and negative because they use different benchmarks. For example, they compare a unit’s performance outcome to both its past aspiration level and performance of referent units (Cyert and March, 1963). If a performance outcome exceeds (falls behind) these benchmarks, managers classify it as a success (failure). These benchmarks, however, do not always agree with each other, producing inconsistent feedback. For example, when a unit’s performance outcome exceeds (falls below) its corresponding aspiration level but falls below (exceeds) the performance of referent units, managers consider this unit to be both successful and unsuccessful. This inconsistency in performance feedback, in turn, makes it difficult for managers to learn from it and move aspiration levels in the direction of their learning (Huber, 1991; Joseph and Gaba, 2015; Kim et al., 2015).

This study thereby introduces performance feedback consistency as a boundary condition for learning from performance feedback in the adjustment of aspiration levels. Specifically, it asks the following question: How does performance feedback consistency influence the impact of attainment discrepancy (i.e. prior performance minus prior aspiration level) and social comparison (i.e. prior performance minus prior performance of “comparable” others) on the current aspiration level? This study answers this question by hypothesizing that past aspiration level, attainment discrepancy, and social comparison positively influence the current aspiration level and that the positive effects of past attainment discrepancy and social comparison become stronger in the case of consistent performance feedback. The empirical test of these hypotheses within the context of 1095 stores of a grocery retailer over 60 quarters shows that previous aspiration level and attainment discrepancy positively influence the current aspiration level and that the effect of social comparison changes across the cases of consistent and inconsistent feedback.

This study makes theoretical, empirical, and practical contributions. On the theoretical front, this article’s investigation of the role of performance feedback consistency in the adjustment of aspiration levels introduces real-world complexity into aspiration theory (Cyert and March, 1963). Also, unlike the literature (Mezias et al., 2002), this study hypothesizes a positive rather than a negative impact of social comparison on aspiration levels as they move in the direction of managerial learning from past performance feedback including the feedback coming from the social comparison (Cyert and March, 1963).

Relatedly, the findings of this study reconcile the mixed evidence for the effect of social comparison on the current aspiration level. Empirical evidence demonstrates that managers either do not adjust aspiration levels in the direction of their learning from the social comparison (Washburn and Bromiley, 2012) or adjust aspiration levels in the opposite direction of their learning from it (Mezias et al., 2002). This study addresses these changing effects of social comparison on aspiration levels. To this end, it proposes and finds that the impact of social comparison is moderated by and thus contingent upon the consistency of performance feedback.

In addition to its contribution to the literature on aspiration levels, this study contributes to the field of administrative sciences and general management. First, given the scarcity of research on setting performance expectations (Schleicher et al., 2018), this study shows how and when managers integrate different types of performance feedback into their performance expectations in the form of aspiration levels. Second, given that the concept of bounded rationality predicts that managers do not process all available performance information (Rahman and De Feis, 2009), this study pinpoints social comparison as the overlooked component of performance information. Third, given that managers proactively seek feedback in today’s business landscape, this study develops two propositions that the literature on feedback seeking behavior (Lam et al., 2017) can investigate.

On the empirical front, this study substantially increases the amount of cumulative evidence on the antecedents of aspiration levels. Accumulating evidence is essential given the scarcity of current evidence that merely comes from a management simulation exercise (Lant, 1992) and units of a bank (Mezias et al., 2002) or a car manufacturer (Washburn and Bromiley, 2012). Given the shortage of past evidence, this article investigates the extent to which previously examined relationships hold in a new empirical context and replicates findings of the previous research (Lant, 1992; Mezias et al., 2002) that prior aspiration level and attainment discrepancy positively increase the current aspiration level.

The findings of this study also have managerial implications. The results suggest that organizational attention (Ocasio, 1997) has an inward focus in the case of inconsistent feedback as managers solely consider a performance outcome relative to an aspiration level under this condition. Besides, the results suggest that attaining “stretch” goals in the long term requires their avoidance in the short term. Another managerial implication of this manuscript concerns transformational leadership (Ng et al., 2016) and change (Bennett et al., 2018). Specifically, the findings of this study suggest that transformational leaders can trigger change through negative attainment discrepancy rather than negative social comparison.

The next section provides an overview of the literature on aspiration levels to highlight the gap that motivates this study. The following section introduces the hypotheses. The section on methods discusses the study context, sample, variables, and method of analysis. The remaining sections elaborate on the findings and discuss their limitations and implications for both practice and future research.

Literature review

There are two main streams of research on aspiration levels. The first stream, which dominates the literature, studies the outcomes of aspiration levels. Specifically, it examines firm-level responses to the performance–aspiration level gap (Gavetti et al., 2012). Corporate lobbying (Rudy and Johnson, 2016), durability of assets (Souder and Bromiley, 2012), expansion of assets (Audia and Greve, 2006), cooperation (Krishnamurthy et al., 2003), efficiency of capital allocation (Arrfelt and Wiseman, 2013), corporate illegality (Bromiley and Harris, 2014; Mishina et al., 2010), growth (Greve, 2008), number of research and development alliances (Tyler and Caner, 2015), cost reduction (Bromiley and Washburn, 2011), and acquisition (Kim et al., 2015) are some of the examples of responses that this stream investigated.

The abundance of work in this first stream has produced five specialized sub-streams. The first sub-stream switched the focus from financial to nonfinancial aspirations such as size (Greve, 2008), new product introduction (Tyler and Caner, 2015), and analyst forecasts (Gentry and Shen, 2013). The second sub-stream teased out the effects of historical and social aspiration levels relative to performance (Kim et al., 2015) and studied the impact of their congruence (Joseph and Gaba, 2015). The third sub-stream investigated various moderators of the aspiration–outcome relationship such as variance in performance (Kim et al., 2015), stock options and ownership (Alessandri and Pattit, 2014), and historical temporal orientation (Souder and Bromiley, 2012). The fourth sub-stream studied prioritization of multiple aspirations (Blettner et al., 2015; Boyle and Shapira, 2012; Chen and Miller, 2007; Greve, 2008; Nielsen, 2014; Washburn and Bromiley, 2012). The final sub-stream studied different empirical specifications of the aspiration–outcome relationship (Bromiley and Harris, 2014; Washburn and Bromiley, 2012).

Despite its benefits, the extensive specialization of the first stream limited the development of the second stream that investigates the drivers of aspiration levels. Consequently, the second stream is comprised of three published studies (Lant, 1992; Mezias et al., 2002; Washburn and Bromiley, 2012) that test the first three hypothesized relationships in Figure 1. Within this stream of research, Lant (1992) took the lead to empirically examine the claims of Cyert and March (1963). Specifically, she tested the impact of past aspiration level and attainment discrepancy, which she defined as the difference between prior performance and previous aspiration level, on the current aspiration level with the data collected from a management simulation exercise. Consistent with her hypotheses, she found evidence for their positive effects.

Figure 1.

Theoretical model: A two-way interaction effect.

Ten years later, Mezias and his associates not only replicated the findings of Lant (1992) in a corporate setting but also introduced social comparison, which they defined as the difference between prior performance and prior performance of “comparable” others, into the literature (Mezias et al., 2002). They hypothesized and found a negative effect of social comparison on the current aspiration level by assuming that “decision makers in organizations will expect to observe similar performance levels among those in the same comparison group” (Mezias et al., 2002: 1285). Given the small economic significance of the effect of social comparison, they, however, concluded that past aspiration level and attainment discrepancy are the most important predictors of the current aspiration level.

Ten years after Mezias et al. (2002) introduced social comparison into the literature, Washburn and Bromiley (2012) compared different model specifications to predict aspiration levels and examined the impact of prior aspiration, prior performance, and prior social comparison when prior performance was either above or below the level of prior social comparison without proposing specific hypotheses for the sign of their effects. They found that previous aspiration level positively influenced the current aspiration level only when prior performance was above social comparison and that prior performance positively influenced the current aspiration level only when prior performance was below social comparison. They did not find any significant evidence for the effect of social comparison, contradicting the finding of Mezias et al. (2002).

Although these original studies have started the long overdue conservation on the antecedents of aspiration levels, the evidence that they provide is still insufficient and partially inconsistent. Also, these studies examine a basic process of aspiration level adaptation and overlook the possibility of inconsistent performance feedback that derives from using different performance benchmarks such as an aspiration level and performance outcomes of “comparable” others. For example, when a performance outcome is simultaneously higher than the corresponding aspiration level but lower than the performance of “comparable” others, managers frame it as both success and failure (Lant, 1992), producing inconsistent performance feedback that is both positive and negative. The resulting inconsistency of performance feedback is problematic because it impairs the ability of managers to learn from it (Huber, 1991; Joseph and Gaba, 2015; Kim et al., 2015) and adjust aspiration levels accordingly.

Thus, this study investigates the role of performance feedback consistency in the adjustment of aspiration levels. In the process, it replicates past research on the antecedents of aspiration levels. Also, it proposes a new sign for the impact of social comparison and documents how its effect differs across the cases of consistent and inconsistent feedback to reconcile its mixed effects.

Hypotheses

Previous aspiration level, attainment discrepancy, and social comparison (Lant, 1992; Mezias et al., 2002) predict the current aspiration level which is “the smallest outcome that would be deemed satisfactory by the decision maker” (Schneider, 1992: 1053).

Specifically, the previous aspiration level positively influences the current aspiration level because managers anchor the current aspiration level on the previous aspiration level rather than on past performance level. Managers give primacy to the previous aspiration level in the adjustment of aspiration levels (Greve, 1998; Lant, 1992) because of inertia, fundamental attribution error, and backward-looking organizational routines (Greve, 2002). They thereby partially and incrementally adjust the current aspiration level to the past performance from the anchor provided by the past aspiration level, producing a positive relationship between the previous and the current aspiration level. Thus, as the past aspiration level increases (decrease), so does the current aspiration level.

Hypothesis 1: The aspiration level in the previous period positively influences the current aspiration level.

In addition to past aspiration level, prior attainment discrepancy and social comparison influence the level of current aspiration. Managers compare prior performance to previous aspiration level or performance of “comparable” others (Mezias et al., 2002). When such comparison shows that a performance outcome is above aspiration level or performance of “comparable” others, managers frame it as a success and conclude that performance feedback is positive. However, when they realize that a performance outcome is below aspiration level or performance of “comparable” others, they frame it as a failure and conclude that performance feedback is negative (Levinthal and March, 1993).

In general, managers adjust the current aspiration level in the direction of their experience (Cyert and March, 1963), prior payoff realization (Borgers and Sarin, 2000) and to the level of achievement (March and Simon, 1958). They, thus, adjust the current aspiration level upward (downward) when they frame the past performance outcome as a success (failure). This type of adjustment, in turn, produces a positive relationship between past attainment discrepancy/social comparison and the current aspiration level. Thus:

Hypothesis 2: Attainment discrepancy in the previous period positively influences the current aspiration level.

Hypothesis 3: Social comparison in the previous period positively influences the current aspiration level.

This manuscript’s hypothesized positive effect of social comparison is different from the negative impact hypothesized by Mezias et al. (2002). They argue for a negative impact because they assume that “decision makers in organizations will expect to observe similar performance levels among those in the same comparison group” (Mezias et al., 2002: 1285). However, decision makers have to evaluate and rank order the performance of organizational units (Fisher et al., 2002). If this were not true, we would not observe the existence of organizational budgeting processes that set different aspiration levels for similar organizational units (Rossi, 2014).

Another reason for why this study does not predict a negative relationship between social comparison and the current aspiration level is that organizational decision makers have more and better information about the performance of their organizational units than outside decision makers (Matvos and Seru, 2014). Unlike outside decision makers, organizational decision makers can collect and analyze performance data from any of their organizational units at any depth, breadth, form, and cadence that they desire. Because of such direct access to unit performance data, organizational decision makers do not have to indirectly infer the performance of a unit from the performance of “comparable” units by expecting “comparable” units to perform similarly.

Moreover, adjusting aspiration levels of socially underperforming units upward and socially over-performing units downward as suggested by Mezias et al. (2002) impairs organizational performance. In particular, this discrepancy reduction logic reduces both variance and mean of aggregate organizational performance over time, impairing competitive advantage (March, 1991). Also, expecting similar performance levels among comparable units requires decision makers to overinvest in a socially underperforming unit while underinvesting in a socially outperforming unit (Bardolet et al., 2011). However, such investment behavior is unlikely because it contradicts past performance.

Because of these reasons, this study does not expect organizations to move aspiration levels of their units in the opposite direction of these units’ comparative performance to ensure performance similarity among them. Actually, rather than seeking social performance similarity, organizational units “fight for dominance at the top of the rank hierarchy” (Kuhnen and Tymula, 2012: 94). Consistent with this inclination of organizational units, empirical evidence shows that discrepancy reduction logic applies merely to relationships between rivals rather than to similar units of an organization (Armstrong and Collopy, 1996).

Although hypotheses 2 and 3 postulate that attainment discrepancy and social comparison positively influence the current aspiration level, the strength of their hypothesized effects depends on performance feedback consistency. Performance feedback is not always consistent due to the disagreement between attainment discrepancy and social comparison (Joseph and Gaba, 2015; Nielsen, 2014). For example, a performance outcome that is above (below) an aspiration level but below (above) the performance of “comparable” others leads to inconsistent feedback because managers classify the very same performance outcome as both a success and a failure under this condition.

The inconsistency of performance feedback, in turn, produces ambiguity and noise (Lant and Mezias, 1992), impairing the managerial ability to learn from it and adjust aspiration levels in the direction of experience and learning (Huber, 1991; March and Olsen, 1975). The same inconsistency also enlarges the variance of performance feedback which, in turn, makes it tough to learn from it (Kim et al., 2015). Supporting this, Lant et al. (1992) concluded that while the high variance of organizational performance in turbulent computer software industry impaired learning from it, the stability of organizational performance in the furniture industry had the opposite effect of facilitating learning from it.

Thus, this study expects attainment discrepancy and social comparison to produce a stronger positive effect on the current aspiration level when performance feedback coming from both attainment discrepancy and social comparison is consistent, leading managers to unambiguously frame the past performance outcome as either success or failure.

Hypothesis 4: Performance feedback consistency moderates the relationship between attainment discrepancy in the previous period and the current aspiration level such that the relationship is stronger when feedback is consistent rather than inconsistent.

Hypothesis 5: Performance feedback consistency moderates the relationship between social comparison in the previous period and the current aspiration level such that the relationship is stronger when feedback is consistent rather than inconsistent.

Figure 1 depicts the hypothesized relationships. The dotted lines represent the impact of the moderator.

Methods

Study context and sample

This study investigated the hypotheses within the context of retailing, following the extant literature (Mezias et al., 2002; Washburn and Bromiley, 2012). Data come from a European grocery retailer where store managers that oversee stores directly report to district managers that manage a geographic market containing multiple stores. District and store managers together set monthly store aspiration levels regarding sales revenue and then work to achieve them. The corporate office of this retailer keeps a repository of these monthly aspiration levels (i.e. targeted sales revenue) and the corresponding performance outcomes (i.e. actual sales revenue) for budgeting, control, and evaluation purposes. This repository also contains information about the identities of managers and the category that a particular store belongs to in line with the firm-specific classification system. The data that this study investigates come from this repository and contain unbalanced monthly data spanning five fiscal years on aspiration and performance levels of 1095 stores, resulting in 35,944 observations.

Data provide several advantages to test the theory. First, data are at store-time level and thus consistent with theory. The theory of Cyert and March (1963) on aspiration levels takes the organization rather than organizational participants as its basic unit. Thus, this study examines data at the level required by the theory and looks at branches of an organization over time following the current practice in all field studies on antecedents of aspiration levels (Mezias et al., 2002; Washburn and Bromiley, 2012). Second, by covering 60 periods, data account for the fact that learning from the past takes time. Third, data correctly identify the social comparison group of a focal store from the perspective of the studied organization. The choice of referents is important and controversial (Shin, 2016) because researchers draw on assumptions and theories to define reference groups. However, in this study, “comparable” others are defined by the studied organization which classifies its stores into predetermined categories for comparison purposes.

Variables

The dependent variable of this study is the current aspiration level. With the exceptions of Mezias et al. (2002) and Washburn and Bromiley (2012), published work has used “factors that influence aspirations in place of actual measures of aspirations” (Bromiley and Harris, 2014: 340). It is, however, important to use internal and direct measures of aspiration levels because this way of measuring aspiration levels reduces measurement error and increases the validity of findings. Thus, like Mezias et al. (2002) and Washburn and Bromiley (2012) that directly measured the current aspiration level through internal performance targets of the studied organization given that “the level of an aspiration is the target performance” (Shinkle, 2012: 422), this study defines the current aspiration level as the sales revenue target of a store in the current period. The studied stores seek to perform above sales revenue targets that are jointly set by district and store managers, making these targets actual aspiration levels of studied stores.

Performance feedback consistency is the moderating variable. The extant literature views performance feedback consistency as a function of the congruence between attainment discrepancy and social comparison (Audia and Brion, 2007; Baum et al., 2005; Joseph and Gaba, 2015). Following this view, this study considers performance feedback to be consistent when attainment discrepancy and social comparison jointly point out success or failure. Consequently, this study operationally defines performance feedback consistency through a dummy variable that equals 1 when both attainment discrepancy and social comparison are either positive, clearly pointing out success, or negative, clearly pointing out failure, and is set equal to 0 otherwise. Measuring performance feedback consistency through a dummy variable is in line with the theory on aspiration levels because the theory postulates that aspirations are heuristics used by boundedly rational managers to turn continuous performance outcomes into dichotomous cases of success or failure (March and Simon, 1958).

In the studied organization, inconsistency between attainment discrepancy and social comparison presents an actual challenge for decision makers in charge of setting aspiration levels. Specifically, managers seek to set the right aspiration levels for the stores under their supervision to earn financial rewards. Thus, they seek all types of performance feedback. However, inconsistent performance feedback due to the discrepancy between attainment discrepancy and social comparison exacerbates their ability to effectively set aspiration levels, increasing the validity of the definition of the moderating variable.

Method of analysis

This study investigated the hypotheses via two-way fixed-effects estimator with robust standard errors because of the findings of several specification tests. First, the significance of Hausman test and Mundlak test, which is robust to both serial correlation and heteroskedasticity unlike the Hausman test, suggested that the fixed-effects estimator was appropriate, ruling out random effects model including multilevel modeling (Singer and Willett, 2003). Next, the evidence for time-specific effects, heteroskedasticity, and serial correlation (Drukker, 2003) suggested a two-way fixed-effects model, which controls for both cross-section and time-specific effects, and heteroskedasticity-robust standard errors that allow intragroup correlation. Also, Fisher’s test checked for a unit root because of investigating many periods and rejected the null hypothesis that all series were nonstationary (Merryman, 2005).

This study complemented its aforementioned main analysis with two additional robustness analyses although its principal analysis of a single firm with two-way fixed-effects estimator controls for the firm and industry effects and accounts for unobserved time- or store-specific effects. The first robustness analysis checked and accounted for endogeneity that can derive from this study’s modeling of current aspiration levels as a function of lagging aspiration levels. This type of modeling can potentially bias fixed-effects estimators. Although the resulting bias is not expected (Baltagi, 2013) in this study because of its investigation of more than 30 time periods (Attanasio et al., 2000; Judson and Owen, 1999), the first robustness analysis still reestimated the full model, model 2 of Table 2, with the bias-corrected least-squares dummy variable estimator (Bruno, 2005).

To this end, it used the Blundell–Bond estimator with no intercept and approximated the bias of the fixed-effects estimator up to order of 1/NT² (Bruno, 2005). Model 3 of Table 2 represents the results of the Blundell–Bond estimator that accounts for potential endogeneity and the resulting bias in coefficient estimates. Model 3 presents only bias-corrected estimates of the coefficients without reporting standard error estimates as it does not report and assess statistical significance (Bruno, 2005).

This study did not resolve endogeneity through alternative methods such as dynamic panel estimation because its large time series component can lead dynamic panel estimation to produce too many instruments, which, in turn, can impair the efficiency of estimates. Also, the results of dynamic panel estimation are too sensitive to the choice of estimation method used. For example, Blundell and Bond generalized method of moments (GMM) estimator and Arellano and Bond GMM estimator can produce different estimates and significance levels from the same data despite meeting the assumptions of dynamic panel estimation (Baltagi, 2013).

The second robustness analysis accounted for a specific type of correlation that can impair the efficiency of the estimates of the primary analysis. In the studied organization, a given district manager supervises multiple stores. This type of supervision, in turn, can correlate aspiration levels of stores supervised by the same district managers and thereby impair the efficiency of estimates of the primary analysis. Thus, the second robustness analysis reestimated model 2 of Table 2 with two-way cluster-robust standard errors which are robust to not only within-store autocorrelation but also within district manager correlation. In addition to providing estimates of parameters that are identical to those of model 2, model 4 of Table 2 provides the corresponding two-way cluster estimates of robust standard errors.

Results

Table 1 provides the correlation matrix and descriptive statistics. The mean value of variance inflation factors (VIFs) did not exceed 4 across all models of Table 2, and none of the values of VIFs was above the generally accepted threshold value of 10 (Cohen et al., 2003), demonstrating that multicollinearity is not an issue to be addressed.

Table 1.

Means, standard deviations, correlations for variables.

Variables	Mean	SD	1	2	3
1. Current aspiration	620,604.4	587,198.7
2. Previous aspiration	611,983.6	575,221.7	0.9644***
3. Attainment discrepancy	−5066.006	87,777.07	−0.0081	−0.0567***
4. Social comparison	81,014.06	404,128.8	0.7397***	0.7721**	0.1390***

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 2, which does not show time dummies to save space, provides the results of the statistical analysis. Model 1 of Table 2 includes the main effects while model 2 of Table 2 introduces the interaction terms. Two-way fixed-effects estimator with heteroskedasticity-robust standard errors estimated these models with STATA as it, unlike SAS, can provide estimates of the constant term, which is essential in the literature. However, these models’ reported estimates of the R ² in Table 2 derive from SAS as SAS provides accurate R ² estimates for fixed-effects estimator (Park, 2009). Apart from these differences, SAS and STATA provide identical estimates of coefficients and their standard errors. Model 3 and model 4 of Table 2 check the robustness of the previous analyses. While model 3 corrects for the potential bias in the estimates of model 2, which includes all the variables, model 4 uses two-way cluster-robust standard errors to correct for potential bias in the estimates of standard errors reported by model 2 of Table 2. Model 3 does not provide estimates of corresponding standard errors as it merely aims to demonstrate the magnitude of bias in coefficient estimates. In overall, data provide support for hypotheses 1, 2, and 5.

Table 2.

Regression models predicting aspiration levels.

	Model 1	Model 2	Model 3	Model 4
Intercept	137,183.5*** (8397.072)	138,017.1*** (8369.889)
Past aspiration	0.7*** (0.01)	0.7*** (0.01)	0.67	0.7*** (0.01)
AD	0.23*** (0.02)	0.19*** (0.05)	0.14	0.19*** (0.05)
SC	0.01 (0.01)	−0.0005 (0.01)	0.02	−0.0005 (0.01)
PFC	4521.29*** (1349.865)	2556.26* (1290.64)	2583.13	2556.26 (1335.84)
AD × PFC		−0.01 (0.06)	−0.002	−0.01 (0.06)
SC × PFC		0.03*** (0.009)	0.04	0.03*** (0.01)
N	35,944	35944	35,944	35,944
R ²	0.9487	0.9488

Note: Standard errors in parentheses. AD: attainment discrepancy; SC: social comparison; PFC: performance feedback consistency.

*p < 0.05; **p < 0.01; ***p < 0.001.

The significant and positive constant term, which is the average of the value of the fixed effects in model 1 and model 2, suggests that studied stores seek high aspiration levels. Although such high aspiration levels can come from managerial overconfidence and optimism (Lant, 1992) and make it difficult for organizational participants to earn their performance-based rewards (Mezias et al., 2002), they can also improve organizational performance by promoting competitive aggression (Ferrier, 2001). Especially for organizations that promote internal competition, the constant term can be significant and positive.

Data provide support for hypotheses 1 and 2, confirming past findings (Lant, 1992; Mezias et al., 2002). Results show that the previous aspiration level and the attainment discrepancy positively and significantly influence the current aspiration level. Thus, the previous aspiration level serves as an anchor for the current aspiration level and that managers adjust the current aspiration level upward for units that outperform their previous aspiration level and downward for units that underperform their previous aspiration level.

Data do not provide support for hypothesis 3 which predicts that social comparison positively influences the current aspiration level. The coefficient estimate of the impact of social comparison is insignificant in model 1 and model 2 of Table 2, replicating the insignificant finding of Washburn and Bromiley (2012). The insignificant coefficient estimate of the social comparison demonstrates that managers do not increase the current aspiration level of units that outperform “comparable” units and lower the current aspiration level of units that underperform “comparable” units.

The empirical support for hypotheses 1 and 2 and the lack of support for hypothesis 3 suggest that managers solely incorporate their learning from previous aspiration level and attainment discrepancy into their adjustment of the current organizational aspiration level. The lack of support for hypothesis 3 and the insignificant effect of social comparison support the overall finding of Lant (1992) and Mezias et al. (2002) that a model containing the previous aspiration level and attainment discrepancy is the best depiction of aspiration level adaptation. The insignificant effect of social comparison is also consistent with the prediction of the social comparison theory that social comparison becomes vital for self-evaluation only when objective performance data are absent (Festinger, 1954). Given the availability of objective performance data in the studied organization, the social comparison takes a back seat in aspiration level adaptation.

Data do not provide support for hypothesis 4 that predicts a stronger positive relationship between the attainment discrepancy and the current aspiration level in the case of consistent performance feedback. The coefficient estimate of the interaction term is insignificant in model 2 of Table 2. Thus, there is no evidence to suggest that the effect of attainment discrepancy differs between consistent and inconsistent performance feedback contexts.

Despite the lack of support for hypothesis 4, there is support for hypothesis 5 that predicts a stronger positive relationship between past social comparison and the current aspiration level in the case of consistent performance feedback. The interaction term between performance feedback consistency and the social comparison is significant in model 2 of Table 2. The insignificant main effects of social comparison and performance feedback consistency coupled with their significant interaction suggest a cross-over interaction, as depicted by Figure 2, which uses values of social comparison that are one standard deviation higher than and less than its mean value respectively (Aiken and West, 1991). The cross-over interaction depicted in Figure 2 shows that the effect of social comparison becomes positive in the case of consistent performance feedback.

Figure 2.

Two-way interaction effects of social comparison and performance feedback consistency on aspiration level.

Robustness analysis

Additional analyses checked the robustness of the findings and corrected potential biases in both parameter and standard error estimates of the full model, which is model 2 of Table 2. The first supplementary analysis concerned the fixed-effects estimates of parameter values in model 2. These estimates can be biased because within-transformation of fixed-effects estimator contemporaneously correlates the past aspiration level, the lagged dependent variable, with the error term. To address this, model 3 reestimated model 2 with the bias-corrected least-squares dummy variable estimator.

Model 3 of Table 2 reports the corresponding estimates. The only notable differences in coefficient estimates between model 2 and model 3 concern attainment discrepancy, social comparison, and interaction terms. While the coefficient estimate of attainment discrepancy drops from 0.19 in model 2 to 0.14 in model 3, the negative coefficient estimate of social comparison in model 2 that overlooks endogeneity becomes positive in model 3 that accounts for endogeneity. Besides, while the coefficient estimate of the impact of the two-way interaction between attainment discrepancy and performance feedback consistency drops from −0.01 in model 2 to −0.002 in model 3, the coefficient estimate of the impact of the two-way interaction between social comparison and performance feedback consistency increases from 0.03 in model 2 to 0.04 in model 3, providing further support for the robustness of the results of model 2.

The second supplementary analysis concerned the estimates of the standard errors in model 2 of Table 2. These estimates can be biased because of the potential correlation of the aspiration levels of stores supervised by the same district manager. To account for such correlation, this study reestimated model 2 of Table 2 with the two-way cluster-robust standard errors which are robust to both within-store autocorrelation and within district manager correlation. Model 4 provides the corresponding estimates of standard errors. The only notable difference in the estimates of standard errors across two models concerns the two-way interaction between social comparison and performance feedback consistency. Its standard error estimate increases from 0.009 in model 2 to 0.01 in model 4. However, this does not change the significance level of its impact. All in all, results suggest that clustering on both store and district managers does not change findings and provides further evidence for their robustness.

Discussion and conclusion

Implications for theory

This study addressed several gaps in the literature on the origins of aspiration levels. First, it replicated previous work not only in a different industry but also in a different country to ascertain the internal and external validity of previous findings, overcoming the dearth of evidence that comes from a simulation exercise (Lant, 1992) and branches of a US-based bank (Mezias et al., 2002) and a car manufacturer (Washburn and Bromiley, 2012). Second, this study introduced real-world complexity into the literature by examining whether and to what extent performance feedback consistency influenced the impact of the drivers of aspiration levels. Third, this study reconciled the mixed evidence for the effect of social comparison on aspiration level (Mezias et al., 2002; Washburn and Bromiley, 2012) by demonstrating that its effect differs significantly across the cases of consistent and inconsistent feedback.

This study also contributed to the field of administrative sciences and general management through its implications for performance management and bounded rationality. Performance expectation setting is an understudied topic (Schleicher et al., 2018). This study contributed to this topic by showing how different types of performance feedback find their way into performance expectations in the form of aspiration levels across the cases of consistent and inconsistent feedback. This study also refined the role of bounded rationality within the context of aspiration level adaptation. Although the concept of bounded rationality predicts that managers do not process all available performance feedback information (Rahman and De Feis, 2009), it does not specify the type of feedback that will be left out. This article addressed this lack of specification and found out that decision makers reduce performance feedback to attainment discrepancy and neglect other types of feedback that disagree with attainment discrepancy. This finding refines the role that bounded rationality plays for managers who learn from past performance feedback to adjust aspiration levels.

Implications for practice

There are critical managerial implications of this study. First, the results suggest that managers should initially set feasible aspiration levels because of their long-term impact and ride on the resulting successes to increase performance target levels over time. Specifically, given that the impact of past aspiration level is larger than the impact of attainment discrepancy, starting with a high aspiration level keeps subsequent aspiration levels high despite low performance and diverges the current aspiration level from actual past performance. The resulting cleavage sets up organizational members for failure and impairs their commitment and motivation as experienced by Tesla that failed to meet 10 of its “stretch” goals on average per year over five years (Pulliam et al., 2016). Thus, managers should avoid high aspiration levels in the form of “stretch” goals that are counterproductive and inefficacious (Gary et al., 2017).

The results also point out performance feedback consistency as the regulator of the direction of organizational attention (Ocasio, 1997). Specifically, results show that organizational attention has an inward focus in the case of inconsistent feedback. For example, when a unit’s performance exceeds its internal aspiration level but falls short of the performance of “comparable” others, managers frame it as a success and provide positive performance feedback. Alternatively, when performance falls below the aspiration level but exceeds the performance of “comparable” others, managers frame it as a failure and provide negative performance feedback. These results suggest an inward focus in the case of inconsistent feedback.

Another managerial implication of this study concerns transformational leadership (Ng et al., 2016) and change (Bennett et al., 2018; Drew and Wallis, 2014; Levene and Higgs, 2018). Although transformational leaders trigger change through negative performance feedback (Shinkle, 2012), the findings of this study show that one type of negative feedback, negative attainment discrepancy, is better than another type of negative feedback, negative social comparison, to trigger change. According to the results, managers should not compare a unit to a more successful comparable unit to produce change via the resulting unsatisfactory performance. Instead, they should rely on attainment discrepancy to produce performance shortfalls, which, in turn, will trigger the need to change. To this end, managers can follow the premises of Leader–Member Exchange theory (Thompson et al., 2018) and develop customized aspiration levels for followers that remain above their actual performance levels.

Limitations and future research directions

This study provides several new directions for future research. First, researchers should investigate the organizational contingencies for a significant and positive constant term detected in this study because such a term inflates aspiration levels and thereby makes managers appear less successful. To this end, follow-up work can examine this term across cooperative and competitive multidivisional structures (Hill et al., 1992). Specifically, future research can hypothesize a positive constant term for competitive multidivisional structures as competition among units can lead them to seek high aspiration levels to obtain more resources from the parent company. Also, future research can expect an insignificant constant term for cooperative multidivisional structures where divisions may be less pressured to increase their aspiration levels to access more resources at others’ expense.

Second, future research should investigate why performance feedback consistency moderates only the impact of social comparison. Future research can specifically investigate whether performance feedback consistency reduces information load of managers, which, in turn, enables them to learn from social comparison and move aspiration levels in the direction of their learning from it.

Third, follow-up work should replicate this study in different countries to examine the role of national culture in aspiration level adaptation. The theory and evidence for the drivers of aspiration levels derive from the analysis of US firms. This study, however, breaks the mold by being the first study to investigate the theory outside the US business context. Follow-up work should continue to examine aspiration level adaptation in different countries to understand better how culture influences the shape, size, and direction of the impact of past aspiration level, attainment discrepancy, and social comparison. To this end, it can use Hofstede’s cultural scale (Hofstede, 2002).

Fourth, the findings of this study pave the ground for two propositions that the literature on feedback seeking behavior can investigate. Feedback-seeking behaviors refer to the proactive actions of employees to ascertain whether they have met or failed performance standards (Lam et al., 2017). There are “four types of feedback seeking about self and others (i.e., self-positive, self-negative, other-positive, and other-negative) based on their foci (i.e., self or peers) and nature (i.e., positive or negative)” (Gong et al., 2017: 1234).

This study’s conclusion that attainment discrepancy prevails over social comparison suggests that self can reign over peers in terms of foci of feedback-seeking behavior. Thus, future research can investigate whether self-positive and self-negative types of feedback seeking behavior are more prevalent and frequent than other-positive and other-negative types of feedback seeking. Also, the finding that the effect of social comparison significantly changes when it agrees with attainment discrepancy can lead future research to investigate whether the impact of other feedback seeking behavior changes depending on its consistency with self-feedback seeking behavior.

There are several limitations of this study, which as a by-product, create additional research opportunities. First, this study defined performance feedback consistency merely in terms of the congruence between historical and social performance feedback. However, such consistency also depends on the congruence of feedback provided by different levels of corporate hierarchy. In practice, different levels of corporate hierarchy can provide different performance reviews of a store depending on their responsibility and priorities. For example, store and corporate managers analyze managerial issues at business and corporate level, respectively. Because of using different levels of analysis, they might reach different conclusions about the success or failure of a store. Thus, future research should expand the definition of performance feedback consistency to include congruence of feedback coming from different levels of corporate hierarchy (Gaba and Joseph, 2013).

Second, this study considers only internal units of the studied organization as “comparable” others by following the extant literature (Mezias et al., 2002; Washburn and Bromiley, 2012) and thereby solely investigates internal comparison. However, organizational units can compare their performance to the performance of “comparable” outside organizations and thereby carry out an external comparison to adjust their aspiration levels. Such internal and external comparison can move aspiration levels in opposite directions, potentially resulting in the detected insignificant main effect of social comparison. Thus, future research should account for the impact of external social comparison to better explain the effect of internal social comparison on aspiration levels.

Conclusion

Although Cyert and March developed the theory on aspiration levels in 1963 and hypothesized about their drivers and outcomes, the literature has focused on the latter (Gavetti et al., 2012; Shinkle, 2012). The extant literature’s such disproportionate focus, in turn, has limited the investigation of the antecedents of aspiration levels to three published studies (Lant, 1992; Mezias et al., 2002; Washburn and Bromiley, 2012).

These studies tested the hypothesis of Cyert and March (1963) that the current aspiration level adapts to past performance feedback and provided useful and preliminary insights. However, being original studies, they did not examine how adapting aspiration levels to past performance changed across the cases of consistent and inconsistent performance feedback. The lack of work on performance feedback inconsistency was problematic because learning from inconsistent feedback is not only a managerial reality but also an impediment to adjusting aspiration levels to past performance. This study addressed this problem and thereby investigated the aspiration level adaptation across the cases of consistent and inconsistent feedback. It found that past aspiration level and attainment discrepancy positively influenced the current aspiration level despite inconsistent performance feedback and that the impact of past social comparison became positive in the case of consistent performance feedback.

Footnotes

Declaration of conflicting interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Burak Cem Konduk

References

Aiken

West

(1991) Multiple Regression: Testing and Interpreting Interactions. Newbury Park: Sage.

Alessandri

Pattit

(2014) Drivers of R&D investment: the interaction of behavioral theory and managerial incentives. Journal of Business Research 67(2): 151–158.

Armstrong

Collopy

(1996) Competitor orientation: effects of objectives and information on managerial decisions and profitability. Journal of Marketing Research 33(2): 188–199.

Arrfelt

Wiseman

(2013) Looking backward instead of forward: aspiration-driven influences on the efficiency of the capital allocation process. Academy of Management Journal 56(4): 1081–1103.

Attanasio

Picci

Scorcu

(2000) Saving, growth, investment: a macroeconomic analysis using a panel of countries. Review of Economics and Statistics 82(2): 182–211.

Audia

Brion

(2007) Reluctant to change: self-enhancing responses to diverging performance measures. Organizational Behavior and Human Decision Processes 102(2): 255–269.

Audia

Greve

(2006) Less likely to fail: low performance, firm size, and factory expansion in the shipbuilding industry. Management Science 52(1): 83–94.

Baltagi

(2013) Econometric Analysis of Panel Data. Hoboken: John Wiley &Sons.

Bardolet

Fox

Lovallo

(2011) Corporate capital allocation: a behavioral perspective. Strategic Management Journal 32(12): 1465–1483.

10.

Baum

JAC

Rowley

Shipilov

, et al. (2005) Dancing with strangers: aspiration performance and the search for underwriting syndicate partners. Administrative Science Quarterly 50(4): 536–575.

11.

Bennett

McCracken

O’Kane

(2018) ‘Engage for change’: the development of a Radical Change Engagement model. Journal of General Management 43(3): 115–125.

12.

Blettner

, et al. (2015) Adaptive aspirations and performance heterogeneity: attention allocation among multiple reference points. Strategic Management Journal 36(7): 987–1005.

13.

Borgers

Sarin

(2000) Naïve reinforcement learning with endogenous aspirations. International Economic Review 41(4): 921–950.

14.

Boyle

Shapira

(2012) The liability of leading: battling aspiration and survival goals in the jeopardy! Tournament of champions. Organization Science 23(4): 1100–1113.

15.

Bromiley

Washburn

(2011) Cost reduction vs innovative search in R&D. Journal of Strategy and Management 4(3): 196–214.

16.

Bromiley

Harris

(2014) A comparison of alternative measures of organizational aspirations. Strategic Management Journal 35(3): 338–357.

17.

Bruno

GSF

(2005) Estimation and inference in dynamic unbalanced panel-data models with a small number of individuals. Stata Journal 5(4): 473–500.

18.

Chen

Miller

(2007) Situational and institutional determinants of firms’ R&D intensity. Strategic Management Journal 28(4): 369–381.

19.

Cohen

West

, et al. (2003) Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. Mahwah: Lawrence Erlbaum.

20.

Cyert

March

(1963) A Behavioral Theory of the Firm. Englewood Cliffs: Prentice-Hall.

21.

Drew

Wallis

(2014) The use of appreciative inquiry in the practices of large-scale organisational change. Journal of General Management 39(4): 3–26.

22.

Drukker

(2003) Testing for serial correlation in linear panel-data models. Stata Journal 3(2): 168–177.

23.

Ferrier

(2001) Navigating the competitive landscape: the drivers and consequences of competitive aggressiveness. Academy of Management Journal 44(4): 858–877.

24.

Festinger

(1954) A theory of social comparison processes. Human Relations 7(2): 117–140.

25.

Fisher

Maines

Peffer

, et al. (2002) Using budgets for performance evaluation: effects of resource allocation and horizontal information asymmetry on budget proposals, budget slack, and performance. The Accounting Review 77(4): 847–865.

26.

Gaba

Joseph

(2013) Corporate structure and performance feedback: aspirations and adaptation in m-form firms. Organization Science 24(4): 1102–1119.

27.

Gary

Yang

Yetton

, et al. (2017) Stretch goals and the distribution of organizational performance. Organization Science 28(3): 395–410.

28.

Gavetti

Greve

Levinthal

(2012) The behavioral theory of the firm: assessment and prospects. Academy of Management Annals 6(1): 1–40.

29.

Gentry

Shen

(2013) The impacts of performance relative to analyst forecasts and analyst coverage on firm R&D intensity. Strategic Management Journal 34(1): 121–130.

30.

Gong

Wang

Huang

, et al. (2017) Toward a goal orientation–based feedback-seeking typology: implications for employee performance outcomes. Journal of Management 43(4): 1234–1260.

31.

Greve

(1998) Performance, aspirations, and risky organizational change. Administrative Science Quarterly 43(1): 58–86.

32.

Greve

(2002) Sticky aspirations: organizational time perspective and competitiveness. Organization Science 13(1): 1–17.

33.

Greve

(2008) A behavioral theory of firm growth: sequential attention to size and performance goals. Academy of Management Journal 51(3): 476–494.

34.

Hill

Hitt

Hoskisson

(1992) Cooperative versus competitive structures in related and unrelated diversified firms. Organization Science 3(4): 501–521.

35.

Hofstede

(2002) Dimensions do not exist—a reply to Brendan McSweeney. Human Relations 55(11): 1355–1361.

36.

Huber

(1991) Organizational learning: the contributing processes and the literatures. Organization Science 2(1): 88–115.

37.

Joseph

Gaba

(2015) The fog of feedback: ambiguity and firm responses to multiple aspiration levels. Strategic Management Journal 36(13): 1960–1978.

38.

Judson

Owen

(1999) Estimating dynamic panel data models: a guide for macroeconomists. Economic Letters 65(1): 9–15.

39.

Kim

Finkelstein

Haleblian

(2015) All aspirations are not created equal: the differential effects of historical and social aspirations on acquisition behavior. Academy of Management Journal 58(5): 1361–1388.

40.

Krishnamurthy

Bottom

Rao

(2003) Adaptive aspirations and contributions to a public good: generic advertising as a response to decline. Organizational Behavior and Human Decision Processes 92(1-2): 22–33.

41.

Kuhnen

Tymula

(2012) Feedback, self-esteem, and performance in organizations. Management Science 58(1): 94–113.

42.

Lam

Peng

Wong

, et al. (2017) Is more feedback seeking always better? Leader-member exchange moderates the relationship between feedback-seeking behavior and performance. Journal of Management 43(7): 2195–2217.

43.

Lant

(1992) Aspiration level adaptation: an empirical exploration. Management Science 38(5): 623–644.

44.

Lant

Mezias

(1992) An organizational learning model of convergence and reorientation. Organization Science 3(2): 47–71.

45.

Lant

Milliken

Batra

(1992) The role of managerial learning and interpretation in strategic persistence and reorientation: an empirical exploration. Strategic Management Journal 13(8): 585–608.

46.

Levene

Higgs

(2018) Leadership and change implementation in a Russian context. Journal of General Management 44(1): 5–16.

47.

Levinthal

March

(1993) The myopia of learning. Strategic Management Journal 14(S2): 95–112.

48.

Matvos

Seru

(2014) Resource allocation within firms and financial market dislocation: evidence from diversified conglomerates. Review of Financial Studies 27(4): 1143–1189.

49.

March

(1991) Exploration and exploitation in organizational learning. Organization Science 2(1): 71–87.

50.

March

Simon

(1958) Organizations. New York: Wiley.

51.

March

Olsen

(1975) The uncertainty of the past: organizational learning under ambiguity. Journal of Political Research 3(2): 147–171.

52.

Merryman

(2005) XTFISHER: Stata module to compute Fisher type unit root test for panel data. Available at: http://ideas.repec.org/c/boc/bocode/s448201.html (accessed 03 January 2015).

53.

Mezias

Chen

Murphy

(2002) Aspiration-level adaptation in an American financial services organization: a field study. Management Science 48(10): 1285–1300.

54.

Mishina

Dykes

Block

, et al. (2010) Why “good” firms do bad things: the effects of high aspirations, high expectations, and prominence on the incidence of corporate illegality. Academy of Management Journal 53(4): 701–722.

55.

Kee

DMH

Ramayah

(2016) The role of transformational leadership, entrepreneurial competence and technical competence on enterprise success of owner-managed SMEs. Journal of General Management 42(1): 23–43.

56.

Nielsen

(2014) Learning from performance feedback: performance information, aspiration levels, and managerial priorities. Public Administration 92(1): 142–160.

57.

Ocasio

(1997) Towards an attention-based view of the firm. Strategic Management Journal 18: 187–206.

58.

Park

(2009) Linear regression models for panel data using SAS, Stata, LIMDEP, and SPSS. Available at: http://www.indiana.edu/∼statmath/stat/all/panel/panel.pdf (accessed 03 January 2015).

59.

Pulliam

Ramsey

Dugan

(2016) Elon Musk sets ambitious goals at Tesla—and often falls short. Wall Street Journal, 15 August. Available at: https://www.wsj.com/articles/elon-musk-sets-ambitious-goals-at-teslaand-often-falls-short-1471275436 (accessed 09 September 2016).

60.

Rahman

De Feis

(2009) Strategic decision-making: models and methods in the face of complexity and time pressure. Journal of General Management 35(2): 43–59.

61.

Rossi

(2014) Capital budgeting in Europe: confronting theory with practice. International Journal of Managerial and Financial Accounting 6(4): 341–356.

62.

Rudy

Johnson

(2016) Performance, aspirations, market versus nonmarket investment. Journal of Management 42(4): 936–959.

63.

Schleicher

Baumann

Sullivan

, et al. (2018) Putting the system into performance management systems: a review and agenda for performance management research. Journal of Management 44(6): 2209–2245.

64.

Schneider

(1992) Framing and conflict: aspiration level contingency, the status quo, and current theories of risky choice. Journal of Experimental Psychology. Learning, Memory, and Cognition 18(5): 1040–1057.

65.

Shin

(2016) Fair pay or power play? Pay equity, managerial power, and compensation adjustments for CEOs. Journal of Management 42(2): 419–448.

66.

Shinkle

(2012) Organizational aspirations, reference points, and goals: building on the past and aiming for the future. Journal of Management 38(1): 415–455.

67.

Singer

Willett

(2003) Applied Longitudinal Analysis: Modeling Change and Event Occurrence. New York: Oxford University Press.

68.

Souder

Bromiley

(2012) Explaining temporal orientation: evidence from the durability of firms’ capital investments. Strategic Management Journal 33(5): 550–569.

69.

Thompson

Buch

Glasø

(2018) Low-quality LMX relationships, leader incivility, and follower responses. Journal of General Management 44(1): 17–26.

70.

Tyler

Caner

(2015) New product introductions below aspirations, slack and R&D alliances: a behavioral perspective. Strategic Management Journal 37(5): 896–910.

71.

Washburn

Bromiley

(2012) Comparing aspiration models: the role of selective attention. Journal of Management Studies 49(5): 896–917.