Markets usually react instantly to significant and unexpected financial news.  But a New York Federal Reserve analyst says these reactions, rather than being knee-jerk, do take into account other prevailing factors in addition to the news itself.   Linda Goldberg vice president in the Bank's Research and Statistics Group, examines the effect of economic news on asset prices in recent NY Fed staff paper.

Goldberg says that most analyses find that economic news is incorporated very quickly, perhaps within minutes into asset prices and that the effect persists for a measurable amount of time and can spill across national borders.  Announcements that are timely, precise, and contain more information generally generate the greatest effect on asset prices

The impact of news on asset prices is often discussed as though there is some governing rule.  For example that a movement of "x" basis points should result from a surprise in the payrolls announcement to a degree mathematically proportionate to the level surprise.  However, Goldberg found there is little reason to expect that the relationship between economic news and asset prices should be stable over time.

The reason for this is basically context.  At different points in time the expected market reaction can change due to prevailing factors such as changing risk preferences in the economy or the state of that economy as it moves closer to or further from targets. The difference in the effect of a "unit of news" on asset prices can be quite large at different points in time.

Goldberg focused on several announcements with 8:30 a.m. releases.  The table below focuses on one of these, nonfarm payrolls, over an eleven-year interval and examined the effects on asset prices.   The data releases generally had a significant effect on the yields of U.S. Treasuries.  Shown are the two and ten year bonds and she says the effects are clear within minutes of the announcement.

Goldberg found that the biggest market movements generally occur when the market reaction isn't behaving normally.  In other words, if we examine the last 13 years of nonfarm payrolls releases compared to their forecast, we could observe that a certain magnitude of "better-than-expected" data corresponds to a certain direction of movement in bond markets.  In the chart below, the average movement would be the flat black line.  The blue line is the actual reaction, rising into positive territory when bonds are reacting more positively for a given level of "surprise."   In other words, this is a chart of how the correlation between data and market reaction has changed over time.



The high point in the chart was during a period of robust economic growth in 2004 when the Federal Reserve began a series of 25-basis-point rate hikes and the low point was in the fall of 2008, particularly in the period before and after the Lehman bankruptcy.  Goldberg suggests that underlying changes in sentiment among investors can predispose certain assets to react more strongly or negatively than they otherwise might. 

Periods of high risk and uncertainty greatly decrease the tendency to "overdo" the typical reaction.  During these times, bond market yields already have some preexisting benefit from the quintessential "flight-to-safety" buying.  Such an environment may also lead market participants to assume the data isn't necessarily indicative of the true nature of the economy when the uncertainty and risk subside.  Taken together, these suggest a bigger surprise in either direction will have a smaller impact than it historically might during times of risk and uncertainty.  In this case, the "news" (i.e. the economic fundamentals that should build a sense of how the economy is doing) is competing with the need to protect against risk and account for uncertainty.

Contrast that to an environment where data is improving--where risk and uncertainty are subsiding, all other things being equal.  Goldberg suggests this amplifies the reaction to news, because "news" has less by way of uncertainty and risk to compete with.  Phrased another way, the news is more reliably indicative of where the economy is heading AND the trading environment itself is less burdened by the need to account for uncertainty.

This becomes increasingly relevant with the current fiscal debate.  The shutdown and uncertainty over the debt ceiling have certainly created uncertainty for markets.  Not surprisingly, we've seen bond markets trade in an extremely narrow range during the height of fiscal uncertainty and begin to move out of that narrow range as headlines emerge suggesting an end in sight to the shutdown and debt ceiling impasse.

The implication for the near term future is that the single most important piece of news for bond markets--the Employment Situation Report (aka "NFP" or "NFP")--can't help but arrive after the slate has been mostly cleared of the type of uncertainty that had been potentially restraining the previous report (because it can't report until the shutdown is over).   While some uncertainty regarding lingering effects of the shutdown and its effect on confidence will remain, the next payrolls report will be far less vulnerable to the "yeah buts" that existed in September, meaning a big deviation from the forecast could pack a bigger punch for rates.