Organizational slack — the cushion of excess resources a firm holds — carries two rival predictions for innovation. The behavioral theory of the firm views slack as a buffer that funds experimentation and protects risky projects, so slack should foster innovation; agency theory predicts that excessive uncommitted resources invite managerial self-interest and discipline loss, so beyond some point overall slack should harm innovation. Marlin and Geiger (2015) moved this debate to the configurational level, clustering 437 U.S. manufacturing firms on the three classical slack types (available, recoverable, potential) and showing that low-slack configurations innovate least while several distinct moderate-slack configurations innovate equally well (equifinality). Crucially, no firm profile combining high levels of all three slack types appeared in their data, so the agency-theory prediction for excessive overall slack — together with the mechanisms linking specific slack bundles to innovation and the causal ordering between configurations and innovation — was left untested. No study has tested the high-overall-slack configuration prediction on innovation outcomes; aggregate-level evidence is mixed and the mechanism gap persists. The problem is to deliver a causal-configurational account: whether high-overall-slack profiles exist and innovate less than moderate-slack profiles, through which mechanisms slack bundles act, and whether slack configurations drive innovation or innovation success instead accumulates slack.

Background

Organizational slack is the cushion of excess resources an organization holds beyond what current operations require (Bourgeois, 1981). Strategy research distinguishes three types of financial slack (Bourgeois & Singh, 1983): available slack, i.e., uncommitted liquid resources, typically measured by ratios such as the current ratio, quick ratio, or working capital to sales; recoverable slack, i.e., excess costs already absorbed in operations that could in principle be recovered or redeployed, typically measured by selling, general and administrative expenses relative to sales; and potential slack, i.e., the firm's capacity to raise resources from outside in the future, typically measured by leverage indicators such as debt-to-equity, debt-to-sales, or debt-to-assets.

Two theoretical traditions make rival predictions about slack and innovation. The behavioral theory of the firm (the view of organizations as coalitions that use excess resources to buffer conflict and absorb the cost of experiments) implies slack should foster innovation: it protects risky projects from short-term performance pressure. Agency theory (the view that managers may pursue their own interests at shareholders' expense when monitoring is weak) implies the opposite at high levels: uncommitted resources create opportunities for self-serving or undisciplined behavior, so excessive slack should reduce innovation. The canonical empirical statement of this tension is Nohria and Gulati (1996), who found an inverted U-shaped relationship — moderate slack is best for innovation — and subsequent studies of single slack dimensions have produced conflicting linear, positive, negative, and curvilinear results (e.g., Geiger & Makri, 2006, on the role of organizational slack in R&D-intensive firms pursuing exploration versus exploitation innovation processes).

Marlin and Geiger (2015) reoriented the debate from single dimensions to configurations, arguing that firms bundle slack types simultaneously and that unidimensional tests cannot capture such bundles. They subjected seven financial slack ratios of 437 U.S. manufacturing firms to factor analysis and cluster analysis — a configurational approach that identifies holistic firm profiles rather than estimating one coefficient per slack type. Six configurations emerged, including 'constrained borrowers' (moderately low on all three slack types), 'depleters' (lowest on all three), 'hoarders' and 'moderate accumulators' (moderately high internal slack combined with moderately high potential slack), and 'spenders/borrowers' (high recoverable slack with low potential slack). Comparing innovation measures — R&D expenditures, patents granted, and patent applications — across configurations while controlling for firm size, diversification, financial performance, and industry innovation levels, they found that the low-slack configurations innovated least, that several distinct higher-slack configurations achieved statistically indistinguishable high innovation (a property called equifinality, meaning different starting profiles reach the same outcome), and that high recoverable slack could compensate for low potential slack. Yet no configuration combined high levels of all three slack types. The observed range therefore confirmed the behavioral-theory-friendly positive pattern only where data existed, and the agency-theory prediction — that firms carrying excessive overall slack innovate less — was never confronted with data.

Problem Statement

Determine how the overall level and configuration of organizational slack affects firm innovation, adjudicating between the behavioral-theory and agency-theory predictions that Marlin and Geiger (2015) could not separate. The problem has three interlinked components, all of which are mandatory:

(1) Configuration test. Do firms whose profiles combine high levels of available, recoverable, and potential slack simultaneously exhibit lower innovation — in innovation inputs such as R&D intensity and in innovation outputs such as patent counts — than firms with moderate-slack profiles, as agency theory predicts, or does the positive behavioral-theory pattern extend across the whole observed range? This includes establishing whether such high-overall-slack profiles empirically exist, how large they are, and, if they are absent from a given sample, bounding the agency-theory prediction by arguing whether that absence leaves the prediction empirically empty; establishing the absence alone does not resolve this component. Here 'high' and 'moderate' are sample-relative: they locate a profile within the distribution of the three slack ratios of the sample studied rather than against fixed absolute cutoffs, so any threshold used to separate high- from moderate-slack profiles must be sample-relative and stated explicitly.

(2) Mechanisms. Through which mechanisms — for example managerial discretion, managerial risk-taking, and resource-allocation processes — do specific bundles of slack types influence innovation? Evidence must link particular configurations (not just aggregate slack) to the posited mediating or moderating processes.

(3) Causal ordering. Do slack configurations drive subsequent innovation, or does innovation success (and the performance it generates) lead firms to accumulate particular slack configurations? Answers must establish an ordering between slack configurations and innovation that survives the reverse-causality and endogeneity concerns inherent in the cross-sectional design of the source study.

Evidence may come from any industry, country, or time period — the source study itself flagged its restriction to U.S. manufacturers as a limitation — provided the three slack types are measured jointly and the study speaks to the stated objective rather than to a single slack dimension in isolation. Single-dimension or aggregate-slack results do not by themselves resolve the configurational question.

References

  1. Marlin, D., & Geiger, S. W. (2015). A reexamination of the organizational slack and innovation relationship. Journal of Business Research, 68(12), 2683-2690. DOI: 10.1016/j.jbusres.2015.03.047. https://doi.org/10.1016/j.jbusres.2015.03.047
  2. Bourgeois, L. J. (1981). On the measurement of organizational slack. Academy of Management Review, 6(1), 29-39. DOI: 10.2307/257138. https://doi.org/10.2307/257138
  3. Bourgeois, L. J., III, & Singh, J. V. (1983). Organizational slack and political behavior among top management teams. Academy of Management Proceedings, 1983(1), 43-47. DOI: 10.5465/ambpp.1983.4976315. https://doi.org/10.5465/ambpp.1983.4976315
  4. Nohria, N., & Gulati, R. (1996). Is slack good or bad for innovation? Academy of Management Journal, 39(5), 1245-1264. DOI: 10.2307/256998. https://doi.org/10.2307/256998
  5. Geiger, S. W., & Makri, M. (2006). Exploration and exploitation innovation processes: The role of organizational slack in R&D intensive firms. The Journal of High Technology Management Research, 17(1), 97-108. DOI: 10.1016/j.hitech.2006.05.007. https://doi.org/10.1016/j.hitech.2006.05.007
  6. Marlin, D., & Geiger, S. W. (2015). The organizational slack and performance relationship: A configurational approach. Management Decision, 53(10), 2339-2355. DOI: 10.1108/MD-03-2015-0100. https://doi.org/10.1108/MD-03-2015-0100
  7. Geiger, S. W., Marlin, D., & Segrest, S. L. (2019). Slack and performance in the hospital industry: A configurational approach. Management Decision, 57(11), 2978-2996. DOI: 10.1108/MD-07-2017-0703. https://doi.org/10.1108/MD-07-2017-0703
  8. Agustí, M., Galán, J. L., & Acedo, F. J. (2022). The effect of slack configurations on company performance from a dynamic perspective. European Management Review, 20(2), 170-187. DOI: 10.1111/emre.12540. https://doi.org/10.1111/emre.12540
  9. Hill, J., & Walton, A. (2021). Impact of organizational slack on innovation outcomes. Journal of Management and Engineering Integration, 14(1), 46-54. DOI: 10.62704/10057/24768. https://doi.org/10.62704/10057/24768
  10. Heubeck, T., & Meckl, R. (2023). Does board composition matter for innovation? A longitudinal study of the organizational slack-innovation relationship in Nasdaq-100 companies. Journal of Management and Governance, 28(2), 597-624. DOI: 10.1007/s10997-023-09687-4. https://doi.org/10.1007/s10997-023-09687-4
  11. Zhang, Y., Sun, Z., & Sun, M. (2022). Unabsorbed slack resources and enterprise innovation: The moderating effect of environmental uncertainty and managerial ability. Sustainability, 14(7), 3782. DOI: 10.3390/su14073782. https://doi.org/10.3390/su14073782
  12. Geiger, S. W., & Marlin, D. R. (2023). Does slack impact innovation productivity? A longitudinal examination. Academy of Management Proceedings, 2023(1). DOI: 10.5465/amproc.2023.11468abstract. https://doi.org/10.5465/amproc.2023.11468abstract
  13. Du, Y., Kim, P. H., Fan, D., & Su, Y. (2025). Zooming in and out for strategic advantage: Slack configurations and R&D investments in high-growth firms. Journal of Business Research, 201, 115729. DOI: 10.1016/j.jbusres.2025.115729. https://doi.org/10.1016/j.jbusres.2025.115729
  14. Mount, M. P., Ertug, G., Kavusan, K., George, G., & Zou, T. (2024). Reeling in the slack: An integrative review to reinstate slack as a central theoretical construct for management research. Academy of Management Annals, 18(2), 473-505. DOI: 10.5465/annals.2023.0087. https://doi.org/10.5465/annals.2023.0087
  15. Depino-Besada, N. I., Vázquez, X. H., Sartal, A., & López-Manuel, L. (2025). An integrative review and research agenda on organizational slack: Addressing assumptions, conflicted theoretical proposals and conclusions. Management Review Quarterly, 76(2), 1201-1249. DOI: 10.1007/s11301-025-00502-y. https://doi.org/10.1007/s11301-025-00502-y
  16. Freitas, A. N., Serra, F. R., Guerrazzi, L., Scazziota, V., & Scafuto, I. C. (2025). Evolution of research on resource slack and future directions. Management Review Quarterly. DOI: 10.1007/s11301-025-00540-6. https://doi.org/10.1007/s11301-025-00540-6
  17. Elston, T., & Zhang, Y. (2025). Ready, willing, and able? Bureaucratic capacity, slack resources, and political control. Journal of Public Administration Research and Theory, 35(4), 452-468. DOI: 10.1093/jopart/muaf021. https://doi.org/10.1093/jopart/muaf021