Monday, February 13, 2017

Paul Krugman's myopic market prediction

I'm not a fan of Paul Krugman. His political bias keeps him from making objective economic analysis but in truth, it's his arrogance and smug attitude that makes me really dislike the guy. As such, I've really enjoyed seeing him eat a little crow regarding his 11/09/16 New York Times blog post. Written the day after Trump was elected president, with the Dow closing at 18,332.74, he says:
"If the question is when markets will recover, a first-pass answer is never...So we are very probably looking at a global recession, with no end in sight. I suppose we could get lucky somehow. But on economics, as on everything else, a terrible thing has just happened."
It didn't take long for Krugman to be proven wrong as the very next day the Dow closed at 18,589.69. Two weeks after the election it closed above 19,000 for the first time in its 120 year history. Then on January 25th it broke above 20,000. As of right now it is at 20,380.81.

I have no idea whether or not the market will remain strong. Given Trump's somewhat anti-free trade noise I'm a little skeptical. Regardless, it's nice to see Krugman quickly proven so completely wrong.

Wednesday, February 1, 2017

Anchoring

Anchoring is a cognitive bias that describes the common human tendency to rely too heavily on the initial piece of information offered (the "anchor") when making decisions.

When making a decision, it is believed that people often make "estimates by starting from an initial value which is adjusted to yield the final answer. The initial value, or starting point, may be suggested by the formulation of the problem, or else it may be the result of a partial computation. Whatever the source of the initial value, adjustments are typically insufficient. That is, different starting points yield different estimates, which are biased towards the initial values."(1)


Judgement Under Uncertainty Study
Amos Tversky and Daniel Kahneman published a famous study on anchoring in the early 1970's. They had participants spin a wheel marked 1 to 100 which unknown to them was rigged to always stop on 10 or 65. They then asked if the percentage of African nations in the United Nations was higher or lower than the number the wheel had landed on. Finally, they asked the participants to estimate the actual percentage.

Tversky and Kahneman found that the number the wheel had landed on had a pronounced effect on the answers the subjects provided. When the wheel landed on 10, the average estimate given by the subjects was 25%. When the wheel landed on 60, the average estimate was 45%.(1)


Real World Example of Anchoring
Sales Negotiations - In a negotiation, the initial price offered can act as an anchor, setting the standard in the mind of the potential buyer.

For example, imagine a negotiation for a used car. In scenario one, the salesman asks for $15,000. The buyer haggles with him and they end up agreeing on $14,500. In scenario two, the salesman asks for $18,000. The two negotiate until settling on a price of $15,500.

In both examples, the sales person anchored the price from which to negotiate in the mind of the buyer. Interestingly, even though the buyer in the first scenario paid less, the buyer in the second scenario is likely to feel he got the better deal as he had negotiated a 'deep discount.'


Prevalence and Susceptibility 
The anchoring bias has been well documented and demonstrated in a variety of domains.(2)  Findings suggest that it is a robust psychological phenomenon and can be difficult to avoid even among those aware of its influence.(3)

Though research findings are somewhat mixed, generally speaking it seems that knowledgeable people are less affected by anchoring. Hence, the more unfamiliar the decision maker is with the problem, the higher the anchoring effect.(2) This seems intuitively true in that a knowledgeable person has access to information with which to judge the plausibility of the anchor.



(1) Judgement Under Uncertainty: Heuristics and Biases: Tversky & Kahneman

(2) A literature review of the anchoring effect

(3) A New Look at Anchoring Effects: Basic Anchoring and Its Anteceents

CIA: Biases in Estimating Probabilities

Anchoring: Wikipedia