How do prudence bias and availability bias work against each other in CME, and which one tends to dominate?

Excellent observation — you've identified a genuine tension in the behavioral bias framework. Prudence and availability can indeed push in opposite directions, and understanding which dominates depends on the context.

The Interaction:

The answer: it depends on whether we're talking about the analyst's private belief or their published forecast.

Private beliefs — Availability tends to dominate:

Internally, the analyst's mental model is heavily colored by recent dramatic events. After a crash, crash scenarios feel more probable. After a boom, continued prosperity feels inevitable. This is availability bias at work — the ease of recalling vivid recent events distorts probability estimates.

Published forecasts — Prudence tends to dominate:

But when the analyst must commit to a public number that will be evaluated by peers, clients, and supervisors, prudence kicks in. Extreme forecasts invite scrutiny. If you publish a -5% equity return forecast and the market returns +12%, your career suffers far more than if you published +7% and the market returned +12%. The asymmetric career incentives push published forecasts toward consensus regardless of private views.

Example — Stonebridge Investments After the 2022 Bond Crash:

After 2022's historic bond losses (the worst in decades), Stonebridge's fixed-income team faced this tension:

• Availability bias made them vividly recall the pain of rising rates. Their gut said: 'Bonds are dangerous. Yields could keep rising.'
• Prudence bias said: 'But consensus expects bonds to recover. If we forecast another bad year and bonds rally, we'll look foolish.'

The result? Privately, the team was more bearish than their published CME suggested. They published a forecast modestly below consensus (4.1% vs. consensus 4.5%) rather than the -2% their availability-colored analysis implied.

Net Effect on CME Quality:

The two biases create a specific pattern:

1. After dramatic events: Availability inflates the perceived probability of recurrence, but prudence prevents the forecast from fully reflecting this view → forecasts are somewhat sticky but nudge toward the recent experience
2. During calm periods: Availability makes extreme events seem implausible, and prudence reinforces this by discouraging extreme forecasts → forecasts cluster near recent trends, understating tail risk
3. Overall: The combination produces forecasts that are too anchored to recent experience AND too close to consensus — getting the worst of both biases

Mitigating Both Simultaneously:

1. For availability bias: Use base rates and long-term data rather than recent memory. Ask 'How often has this happened historically?' rather than 'How easily can I imagine it happening?'
2. For prudence bias: Create a culture where extreme forecasts backed by solid analysis are rewarded, not punished. Separate the messenger from the message.
3. For both: Use structured analytical frameworks (building blocks, scenario analysis with explicit probabilities) that force systematic reasoning rather than relying on intuitive judgment.

Practice behavioral bias questions in our CFA Level III question bank.