Behavioral finance argues that, thanks to various behavioral influences, often there is a discrepancy between market price and intrinsic value. The arguments of behaviouralists rests on two key assumptions:

1. Some investors – they call them noise traders are not rational as their demand for risky assets is influenced by beliefs or sentiments that are not fully supported by fundamentals.
2. Arbitrage operation by rational investors tends to be limited as there are risks associated with it.

Noise Trading:

Many investors trade on pseudo-signals, or noise, and not on fundamentals. As long as these investors trade randomly, their trades cancel out and are likely to have no perceptible impact on demand. True, this happens to some extent because the market is thronged by noise traders who employ different models and, hence, cancel each other out. However, a good portion of noise traders employ similar strategies, as they suffer from similar judgmental biases while processing information. For example:

1. They tend to be overconfident and hence assume more risk
2. They tend to extrapolate past time series and hence chase trends.
3. They tend to put lesser weight on base rates and more weight on new information and hence overreact to news.
4. They follow market gurus and forecasts and act in a similar fashion.

Given the correlated behavior of noise traders their actions lead to aggregate shifts in demand.

Limits to Arbitrage:

One can expect the irrationality of ‘noise traders’ to be countered by the rationality of arbitrageurs as the latter are supposed it be guided by fundamentals and immune to sentiments. However, arbitrage in the real world is limited by two types of risk. The first risk is fundamental. Buying undervalued securities tends to be risky because the market may fall further and inflict losses. The fear of such a loss may restrain arbitrageurs from taking large enough long positions that will push price to fully conform to fundamentals.

The second risk is resale price risk and it arises mainly from he fact arbitrageurs have finite horizons. There are two principal reasons:

1. Arbitrageurs usually borrow money or securities to implement their trades and, therefore, have to pay fees periodically. So they can ill-afford to keep an open position over a long horizon.
2. Portfolio managers are evaluated every few months. This limits tier horizon of arbitrage.

Price Behavior:

Given the substantial presence of noise traders whose behavior is correlated and the limits to arbitrage, investor sentiment does influence prices. In such a market, prices often vary more than what is warranted by changes in fundamentals. Indeed, arbitrageurs may also contribute to price volatility as they try to take advantage of the mood swings of noise traders. For example, when some investors follow a positive feedback strategy that says buy when the price increases and sell when the price decreases, it is no longer optimal for arbitrageurs to counter the actions of noise traders all the time. Instead, they may profit by jumping on the bandwagon themselves for a while. It pays them to buy stocks which excite feedback traders, stimulate price increase, fuel the purchase of other investors, and sell near the top and collect their profits. Likewise, it is profitable for them to sell stock that positive feedback traders dislike, trigger price decreases, induce sales by other investors and buy them back near the nadir. Finally their action would align prices to fundamentals. The effect of arbitrage is to stimulate the interest of other investors and so to contribute to the movement of prices away from fundamentals. Although eventually arbitrageurs sell out and help prices return to fundamentals, in the short run they feed the bubble rather than help it to dissolve.
–