When Policy Becomes Volatility

Most market information arrives as a revision: an earnings estimate changes, a data release shifts a forecast, a probability moves by degrees. The tariff announcement of 2 April 2025 changed the shape of the distribution instead. Risky assets sold off sharply across several sessions; on 9 April, after a ninety-day pause was announced for most countries, the S&P 500 gained 9.5 per cent in one day. The economic outlook had not moved smoothly down and then smoothly up. Markets were repricing branches of policy, which is why the relevant object was jump risk rather than a higher setting of ordinary variance.

Section 01Two kinds of risk

It helps to separate two things that the word volatility runs together. The first is diffusive risk: the continuous, small, roughly symmetric jostling of prices that classical models describe well. The second is jump risk: discrete, large moves of uncertain sign, triggered by events. A scheduled policy decision is almost pure jump risk. The market often knows roughly when the announcement will come, but not what it will say, and the outcome is closer to a branch than to a drift, tariffs imposed or tariffs paused, with a wide gap between the two branches.

Options markets price these two risks differently, and the difference is visible in their prices. Where jumps are feared, implied volatility rises, the distribution priced into options grows fatter tails, the skew steepens because crash protection is bid, and short-dated options become expensive because the event is near. The theoretical home of this intuition is Robert Merton's work on pricing options when returns can be discontinuous: once you admit jumps, the price of an option must include a term for the size and frequency of those jumps, not merely the day-to-day variance. An options surface in front of a policy event is, in effect, the market's quote for a discontinuity it cannot yet sign.

Section 02The transmission chain

Figure 4 follows the announcement as it is repriced, stage by stage, in different languages. It begins as policy uncertainty, the one input a portfolio cannot diversify away because it cannot be forecast. It reaches rates and foreign exchange first, as haven assets are bid and the curve shifts. It hits equity futures next, often as an overnight gap, because policy news tends to arrive when the cash market is closed and the future is the only thing trading. It then deforms the options surface: short-dated implied volatility spikes, the skew steepens, and the term structure can invert as near-term uncertainty exceeds the long-term view. Finally it reaches intraday liquidity, where displayed depth thins exactly when participants most need to trade, a theme this series returns to in its own right.

The cross-asset evidence from April 2025 fits the chain. The Bank for International Settlements, reviewing the episode, noted that the VIX more than doubled in the immediate aftermath of the announcement before retreating below its pre-announcement level, and that the index recovered all of its early-April losses by the end of the same month. A single shock had been repriced through every layer of the market and then, in large part, repriced back. The volatility was real; the lasting change in the level of prices was, in the end, much smaller than the violence of the moves suggested.

price path:   dS/S  =  mu*dt  +  sigma*dW  +  J*dN
                                  (diffusion)   (jump on event arrival dN)

implied vol before a known event  embeds  E[J^2]   # the surface, not a level
realised vol^2  ~  sum_t ( r_t^2 )                  # depends on the actual path

a round trip (down, then up) makes realised vol large
even when start and end are close together

Section 03Volatility is path-dependent

The rebound is the part worth dwelling on. Realised volatility is built from the squares of returns along the path, so a round trip, down roughly twelve per cent and then up roughly nine and a half, generates an enormous quantity of realised volatility even though the start and end points are close. A market that falls and recovers has not been calm; it has been violent twice. This is why a single figure, one move or one index print, cannot describe what happened. The information is in the sequence.

Hedging flows can amplify such paths when dealer positioning makes hedges procyclical, but that mechanism is not measured here and should not be treated as an attribution for the episode. The more defensible result comes from the Bank for International Settlements' later decomposition: it estimated that roughly three-quarters of the S&P 500's rise from the 9 April trough through the end of July reflected positive surprises unrelated to tariffs, including macroeconomic and earnings news. Policy reversals helped offset the initial shock; they did not explain the entire recovery.

A market that falls and recovers has not been calm. It has been violent twice, and a single number cannot say so.

The episode separates three objects that are often collapsed into one word. Before the announcement, option prices represented the market's distribution over possible jumps. During the repricing, liquidity and hedging determined the path through those states. Afterward, realised volatility recorded every move even though the index recovered much of its loss. Policy uncertainty is therefore not adequately described by a volatility level. It is a dated distribution, a market-transmission process, and a realised sequence.

This research is analysis and commentary for general information. It is not investment advice, an offer, or a solicitation, and it contains no price forecasts. Market figures summarise the cited sources; the interpretation is the author's.