Podcast Summary: LSE Public Lectures and Events

Episode: In Praise of Weak Incentives

Guest Speaker: John Roberts (Stanford Graduate School of Business)
Date: March 26, 2009

Overview

This episode features economist John Roberts discussing "In Praise of Weak Incentives." Against the background of the global financial crisis, Roberts challenges the prevailing view that strong, market-like incentives are always beneficial within organizations. He argues that in many key situations—especially when measures of performance are poor, jobs are multi-dimensional, or cooperation and experimentation are critical—weak or even flat incentives can produce better outcomes and prevent catastrophic failures, such as those seen in the financial sector.

Key Discussion Points & Insights

Introduction and Context

• Roberts is introduced by David Metcalfe as an influential economist and author who blends theoretical and practical perspectives, including real-world organizational experience (00:00–01:54).
• The lecture is inspired by the financial crisis, which Roberts attributes largely to the "badly designed strong incentives" that permeated the financial sector (02:35–03:19).

The Danger of Strong, Bad Incentives

• Financial sector failures (subprime mortgages, CDOs, AIG's collapse) are traced to agents responding to single-minded, strong incentives—often commissions or short-term profits—without accountability for quality or long-term risks (03:12–09:40).

  • “People were responding to strong incentives, but strong, badly designed incentives.” – John Roberts (03:12)

• Incentives throughout the chain: Each party (mortgage originators, banks, investors, rating agencies) had strong financial reasons to overlook risk, pass along bad products, or inflate ratings, resulting in systemic disaster (05:00–09:40).

When Weak Incentives Are Superior

Roberts details five contexts where weak incentives inside organizations may outperform strong incentives:

1. Poor Performance Measures (18:33–22:21)

• When performance measures are noisy or biased (e.g., short-term results over long-term value), strong incentives lead to misalignment with true goals and erratic, risk-averse, or destructive behavior (21:07–21:43).
  • "If you pay them for short-term returns, that's what you're going to get.” (21:07)
• Recommendation: Use weak incentives or straight salary when measurement is unreliable.

2. Multi-Tasking Roles (22:22–29:54)

• Jobs often contain multiple objectives, some easier to measure than others (e.g., sales vs. relationship building).
• Strong incentives for measurable tasks result in the neglect of important but hard-to-measure activities. Balance in incentives is essential but costly; thus, weak, broad-based incentives may be preferable (23:13–31:00).
  • "You have to give balanced incentives...if you want your sales staff to be building relationships, what you have to do is not have them be operating just on commission, put them on salary, make them full-time employees." (27:31)

3. Necessity of Cooperation (29:54–33:29)

• Many organizations require collaboration and mutual support, creating "spillovers" of value not easily assigned to individuals.
• Strong individual incentives can undermine collective objectives.
• Recommendation: Weaken direct individual incentives and increase focus on shared or group success.

4. Encouraging Experimentation (33:30–35:36)

• Experimentation often involves short-term loss or risk with uncertain payoffs.
• Strong ties between pay and early results discourage risk-taking; weak incentives allow for beneficial experimentation (35:36).
  • “If you want to motivate experimentation...you shouldn’t reward him for early successes, or certainly not strongly...” (35:36)

5. Managing Disagreement & Authority (35:37–39:51)

• In principal-agent relationships with fundamental disagreement, strong incentives simply motivate the agent to follow their own judgment, not the principal’s instructions.
• Recommendation: Weak or flat incentives, with disciplinary levers (like the threat of firing), keep authority intact.

The Broader Economic Model

• Markets naturally give strong incentives. Firms exist, in part, to allow for the application of desirable weak incentives where market-like strength causes harm (15:11–17:53).
  • “Markets don’t know how to give weak incentives. Markets automatically give strong incentives. But if strong incentives are bad incentives, then what you need is something like the firm.” (15:11)

Notable Quotes & Memorable Moments

• On Corporate Governance Failure:
  "Badly designed, strong incentives are a recipe for disaster. So fine, the answer is design good incentives. Well, that's hard...(11:31)"
• On the Function of Firms:
  “The firm is really usefully viewed as an institution to permit giving weak incentives.” (15:13)
• On CEO Compensation:
  “I think they are given more incentive pay than they need now...It leads them to take risks that I don't think shareholders want taken, because they don't face any downside.” (52:26)
• On Educational Incentives:
  "The Toyota workers don't receive any individual performance incentives...they're motivated by pride in their job, they're motivated by being trusted, by a certain level of monitoring." (44:00)
• Audience laughter at Tottenham football reference:
  “David and I are bound by our love of Hotspurs. Nobody laughed. That’s nice.” (36:11)

Audience Q&A Highlights

[41:15] State/Market Incentives (B–C)

• Q: Does advocacy for weak incentives suggest centralization/government planning?
• A: No, Roberts clarifies he’s offering a counterpoint to years of market-worship and notes that markets and strong incentives work well when measures are sound and consequences direct.

[47:40] Financial Crisis Complexity (E–C)

• Q: Were contract writers simply mistaken or was there deeper agency misalignment in firms?
• A: It’s multi-layered: misaligned incentives, moral hazard, model simplifications, and competitive pressures all contributed.

[51:07] Senior Management (B–C)

• Q: Can senior managers perform if paid mostly flat salaries?
• A: Yes—excessively strong incentives (200x average worker) aren’t needed for alignment or effort, and may encourage excessive risk.

[55:43] The Rise of Market Incentives (E–C/D)

• Q: Why did strong incentives gain so much academic and business traction?
• A: Markets performed well for decades; opinion leaders (CEOs) benefited most, driving adoption and propagation. However, this approach has been pushed too far.

[62:37] History and Heritage of the Idea (F–C/D)

• Q: Do weak incentives have academic heritage?
• A: Yes, going back to Coase and Williamson. Historically, authority (not incentives) was stressed inside firms; only recently have formal models reflected the role of "weak" performance pay and authority structures.

[68:06] Application to Investment Banking (G–C/D)

• Q: Are weak incentives relevant to investment banking, where measures and goals are clear?
• A: If measurement is truly strong and multitasking limited, strong incentives are fine; however, evidence of systemic failures suggests measures may not be as robust as believed.

[72:03] Innovation Gone Wrong? (E–D/C)

• Q: Didn’t strong incentives drive financial "innovation" (e.g., CDOs) that led to the crisis?
• A: Yes. Innovation boomed, but without regulation and oversight, strong incentives became destructive.

Summary Table: When to Use Weak Incentives (Selected Segments & Timestamps)

| Context | Description | Time | Key Point | |---------------------------|------------------------------------------------------------------------------|----------|---------------------------------------------------------------------------------------| | Poor Performance Measures | When output is hard to evaluate accurately | 18:33–22:21 | Strong incentives lead to gaming or risky behavior; use salary/weak incentives. | | Multitasking | Multiple objectives, some hard to measure | 22:22–29:54 | Avoid neglect of important, unmeasured tasks; rely on broad, weak incentives. | | Cooperation | Need for teamwork and spillovers | 29:54–33:29 | Incentivizing only individual outcome undermines group goals. | | Experimentation | Encouraging risky, innovative behavior | 33:30–35:36 | Weak incentives especially early on, then ramp up for implementation/success. | | Disagreement & Authority | Principal and agent differ fundamentally on what should be done | 35:37–39:51 | Strong incentives push agent to ignore principal's wishes; weak incentives restore authority. |

Tone and Style

Roberts’ lecture combines clear economic logic, accessible examples (Toyota, teacher pay, CEOs), humor (Tottenham Hotspurs), and engagement with contemporary events. The tone is candid and reflective, offering nuanced challenges to simplistic pro-market, high-powered-incentive narratives.

Conclusion (39:51)

"Bringing activities inside the firm allows provision of desirably weak incentives. And attempting to recreate market incentives inside firms is thus often a mistake. And it's that attempt...inside the banking system that I lay as the cause of much of the crisis that we're suffering, genuinely suffering through now." – John Roberts (39:51)

The episode offers economists, managers, and policymakers a new lens to reconsider the design of incentive systems, especially after crises rooted in runaway "performance" pay.