Volatility: What Causes Price Swings and Why It Matters (Chapter 20)
Volatility is one of those words that everyone uses but most people think about too simply. Prices went up and down a lot today? Volatile. VIX is high? Volatile. Your crypto portfolio lost 40%? Very volatile.
Harris takes Chapter 20 to explain that volatility is not one thing. It is two things. And the distinction between those two things matters enormously for traders, for regulators, and for anyone trying to understand whether markets are actually working properly.
Fundamental Volatility
Fundamental volatility happens when the actual value of an instrument changes due to new information. A company reports terrible earnings. A frost destroys the Florida orange crop. A government defaults on its bonds. These events change what things are genuinely worth, so prices should change. This kind of volatility is not a bug. It is a feature.
If an unexpected killer frost hits Florida overnight, orange juice futures should open at a much higher price the next morning. Everyone knows about the frost from the morning news. Prices adjust without much trading because the information is public.
But when only a few people know new information, prices change on high volume. Informed traders buy or sell on their private information. Their trading pressure pushes prices toward the new fundamental values. Dealers, who cannot tell informed traders from uninformed ones, respond by adjusting their quotes. This is the adverse selection spread component at work.
Harris identifies several factors that drive fundamental volatility:
Storage costs matter a lot for commodities. Things that are expensive to store tend to have volatile prices because producers and distributors do not hold large inventories. When demand exceeds supply, inventories deplete fast and prices spike. Electricity is the ultimate example. It is essentially impossible to store, so spot electricity prices can be wildly volatile. California experienced this dramatically in 2000-2001 when prices occasionally spiked during periods of high demand.
Fundamental uncertainty drives volatility in tech stocks and emerging markets. When a company’s value depends on research outcomes and markets for products that do not exist yet, even the best-informed traders have very little information. New data or new valuation models cause large price swings. This is why high P/E stocks tend to be more volatile than low P/E stocks. Their prices embed expectations about growth that depend on many uncertain future events.
Leverage amplifies volatility for equities. When a company has a lot of debt, the equity holders absorb most of the variation in asset values. Small changes in the total value of assets produce large changes in equity value because bondholders get paid first.
Political risk creates volatility in sovereign debt, regulated industries, and any instrument whose value depends on government decisions. Wars, nationalization threats, inflation policy, and regulatory changes all create fundamental uncertainty that cannot be diversified away.
Here is an important point Harris emphasizes: expected changes in fundamentals generally do not create volatility. If everyone knows that a seasonal pattern will affect demand, that is already in the price. Only unexpected changes move prices. This is why fully informative prices follow a random walk. If you could predict the next price change, someone would have already traded on that prediction.
Transitory Volatility
Transitory volatility is the other kind, and it is fundamentally different. This is what happens when impatient uninformed traders push prices away from fundamental values. These price changes are temporary because prices eventually revert back.
The simplest form is bid/ask bounce. Market orders buy at the ask and sell at the bid. Prices bounce back and forth between these two prices even when fundamental value has not changed at all. The transaction cost component of the spread is responsible for this bouncing. It is a real cost to traders but has nothing to do with actual changes in value.
Large orders and cumulative order imbalances from uninformed traders also create transitory volatility. When a big pension fund dumps a large block of stock for non-informational reasons (rebalancing, meeting redemptions), the price drops. But the drop is not because the company is worth less. It is because someone needed liquidity and was willing to pay for it. Value traders and arbitrageurs eventually recognize that prices have moved away from fundamentals and trade them back.
Here is the critical connection: transitory volatility and transaction costs are essentially the same thing, viewed from different angles. The price impacts that uninformed traders have on prices are their transaction costs. Those same price impacts are what create transitory volatility. This is why transitory volatility is small in liquid markets and large in illiquid ones.
Why the Distinction Matters
For traders, confusing the two types leads to bad decisions. If you see prices dropping and assume it is fundamental (the company is in trouble), you might sell into what is actually transitory volatility (a big uninformed seller pushing prices down temporarily). You would be selling at exactly the wrong time.
Options traders care about this intensely. Option values depend on volatility, but they really depend on fundamental volatility. Transitory volatility creates noise but eventually reverts. If you price options based on total observed volatility without decomposing it, you will misprice them.
For regulators, the distinction is even more important. Regulators can do essentially nothing about fundamental volatility. If the world genuinely changes, prices should change too. Trying to suppress fundamental volatility would just make prices less informative, which is worse for the economy.
But transitory volatility is something regulators can influence. Market structure decisions, tick sizes, transparency rules, and trader access policies all affect how much uninformed trading impacts prices. High transitory volatility signals that markets are illiquid and might need structural improvements.
When markets crash and everyone demands that regulators “do something,” the first question should be: is this fundamental volatility (the economy genuinely deteriorating) or transitory volatility (panic-driven selling that will reverse)? The appropriate response depends entirely on the answer.
Measuring the Components
How do you tell the two apart? The key is price reversals. Fundamental price changes do not revert. If a company’s factory burns down, the stock price drops and stays down. But transitory price changes do revert. If an uninformed trader pushes the price down, it bounces back when value traders step in.
This means transitory volatility shows up as negative serial correlation in price changes. Increases tend to follow decreases and vice versa. Price reversals are more common than price continuations. The presence of negative serial correlation in a price series is a strong indicator of transitory volatility.
Harris introduces Roll’s serial covariance spread estimator, which uses this insight to estimate bid/ask spreads from trade price data alone, without needing bid and ask quotes. The basic idea: if you observe negative correlation between adjacent price changes, some of that is due to bid/ask bounce. The magnitude of the negative correlation tells you how big the spread is.
The limitation of Roll’s model is that it only captures reversals between adjacent trades. If the reversion takes longer, you need more sophisticated models that can decompose a price series into a random walk component (fundamental) and a mean-reverting component (transitory) over longer horizons.
One subtlety Harris points out: negatively correlated spot prices for perishable commodities might look like transitory volatility but actually reflect fundamental volatility. Monday’s fish price and Tuesday’s fish price are prices for different things (fish delivered on different days). The correlation structure reflects changes in fundamentals across delivery dates, not price reversals in the same asset.
The Bottom Line
Volatility is not inherently bad. Fundamental volatility is the market doing its job, incorporating new information into prices. Transitory volatility is the cost of the market doing its job imperfectly, the friction created by the trading process itself.
Traders need to distinguish between the two to predict future volatility accurately and to know when prices are presenting genuine trading opportunities versus just bouncing around. Regulators need to distinguish between them to know when intervention is warranted and when it would just make things worse.
The chapter is short but dense. And it sets up the next chapter perfectly, because once you understand what volatility is, the natural next question is: how do you actually measure the costs that trading imposes on participants?
Next: Measuring Liquidity and Transaction Costs
This post is part of a series on Trading and Exchanges: Market Microstructure for Practitioners by Larry Harris (Oxford University Press, 2003).