Lower yields ahead of hike

Hi All,

Re: the FOMC announcement today, the market is almost 100% expecting a hike, causing a drop in treasury yields (http://www.marketwatch.com/story/treasury-yields-drop-as-investors-brace-for-fed-rate-hike-2016-12-14).

I’m trying to understand the rationale for the decreased yields, given the rate hike. Admittedly, my understanding of fixed income is not as strong as I’d like it to be but I want to learn and am throwing the question out to the community. Any help would be great.

My thought process is as follows: - A very raw form of calculating yield is coupon rate/market price of bond (assuming flat rate curve). A drop in yield would imply that either coupons have gone down or market price of bond has gone up

Based on the behaviour today, investors are excessively buying treasuries today, drives up the price >>> increases denominator >>> decreases yield.

But just thinking out loud: - Since the rate curve has been adjusted up i.e. future cash flows are being discounted at successfully higher rates, reducing PV of cash flows (reducing the current value/price >>> shouldn’t this be increasing yield based on my simple fraction above) or,

  • Using the YTM angle, since current treasury coupons will now be invested at higher interest rates after the hike, shouldn’t this increase the yields of current bonds or,

Again, my explanations above may be very “by-the-book”. But any help is appreciated on the last 2 questions - or any other comments on my guess of lower yields above.

Lower yields suggests money is flowing in to fixed income in some way. To me, that suggests that people are parking stuff in fixed income, perhaps to reduce duration on other assets (duration is most common in fixed income, but any asset that is interest-rate sensitive effectively has a durations).

There may be some derivatives effects too, just in case people are using synthetic exposures to neutralize commodity or equity effects. If people are reducing exposure, that suggests they buy a bond and short the underlying, which would push money at least temporarily into fixed income (though this would mostly affects the near end of the yield curve).

Since rates going up is generally thought to depress stock and commodity prices, I can see why people might want to neutralize exposure through futures for a day or so.

Plus, higher interest rates over the long term tend to depress growth vs the case where they are low, so I could see some downward pressure on the far end of the yield curve too, although that part is most likely priced in already.

What is weird about the time we live in is that the normal effects of yields (increase in yields -> decrease in stock PVs) is counteracted by the idea that a fed funds rate increase indicates confidence in the economy, which makes people want to buy stocks. So now you have a situation where increase in yields might push stock prices up because of a confidence effect. However, stocks do tend to have interest rate effects that are theoretically on par with 30 or 50 year bonds.

Generally, the idea the Fed has is to increase rates when the economy is growing fast enough that the rate increase won’t interrupt economic recovery, even if it might slow it down a bit. But while there has been a financial assets recovery, the economic growth underlying it is pretty weak. Employment seems to be at a decent level, but wages have only crept up a tiny bit and only in the last quarter or two. Meanwhile job security is still very low so wage workers can’t really plan or save very effectively, and many people have overhanging debt from the great recession whose interest rate is way higher than the rate of wage growth. It’s not as though it’s particularly safe to raise rates now, but neither is it very safe to keep them low forever.

All these feedback loops get confusing, but it’s part of why markets are so fascinating.

This is pretty elementary. You can do some reading on your own. But anyway…

Fed releases a whole set of data on each FOMC announcement date. The most important of this information is the Fed Funds target range. However, they also release some other material information - for instance their guidance for future rate policy decisions (the “dot plot”). It could very well be that they increase the Fed Funds rate and decrease their projections for future rate hikes, like they did in June. The Fed Funds rate is an overnight lending rate and might not influence 10y yields to the same extent as it would short term rates like 3m libor or something.

There is also a basis between Fed Funds rate and Treasury rates. For instance, the Fed could signal that they expect fiscal projects would are expected to have less inflationary pressure than previously anticipated. This would drive down Treasury rates independently from changes to the Fed Funds rate.

Despite all that, I suspect that the market action described in the article is just unrelated to the Fed Funds decision, which has already been indicated strongly. Market changes that happen around the same time as the release of key data do not have to be driven by that data, nor does the article say that this is. If anything, uncertainty causes people to buy safe assets like Treasuries, so market prices do not always have to be based on fundamentals.

It is true that journalistic coverage of market events tends to follow the formula

[Market index changes by X] due to [Some market event that is in the news and possibly relevant.]

In other words, just because the headline says or implies that two things are related doesn’t mean that they actually are related. It’s just that the journalist has a deadline to meet and most people will forget what the journalist said the next day anyway, so why not say something plausible and move on to the next item before you get fired.

The article does not say that the two are related. I think they very purposely word things this way to avoid this implication, even if putting those things in the same article can be seen as suggestive.

"Treasury yields drop as investors brace for Fed rate hike " (they happened at the same time)

“Investors tepidly bought Treasury bonds on Wednesday, pushing yields lower ahead of an important Federal Reserve meeting” (investors bought bonds before the meeting, but not necessarily because of the meeting)

Thanks so much guys! Much appreciated.

All of what I had read was also eluding to yields down because prices up due to demand. But, as you mentioned, bchad, there are some aberrations in the current world we live in - I have always read that if money flows out of one category (bonds) it most likely flows into the other (equity) or vice versa. But everything is pointing up now - and agreed with your points on the underlying variables dont seem to justify the confidence that the markets seem to be pricing in. Also, maybe a tangent (and unrelated to my question on yields) but just the run-up in equities reflects a lot of optimism post-election (lower regulation etc.), further fueling equities. So, the yields could be caused by other factors that I wasnt aware of (similar to derivatives + hedging you mentioned) and hence, wanted to throw the question out there.

Thanks so much for the explanations - just want to make sure my fundamentals are clear.

I did modify my post to say “or imply”, which is what they do more often, but you [Ohai] probably didn’t get to see the change before you posted.

I agree that it is a lot like diplomatic speak, where you make sure to hedge things to the precise degree you want. It’s easier to do in text, but headlines are much harder to get right (it does seem that they did do it here).

Nonetheless, the headlines are often suggestive, so my point is simply to take journalistic headlines with a grain of salt in terms of any causality stated or implied unless you see strong evidence in the body of the article or elsewhere.

Ok, I’m guilty of posting without really reading the article and just taking on faith that the yields dropped this morning.

It looks like it’s the far end of the yield curve that is dropping, 10y and 30y, meanwhile the 2y is climbing. I don’t track these things on a daily basis so can’t really speak to whether this is just noise or some kind of unusual drop (just because the article mentions changes, doesn’t mean that it’s an unusual size), but assuming it is, it is interesting that it would imply money flowing in to the far end of the curve, which has a higher duration, but also tends not to be super highly correlated with Fed funds rate changes. I’d have to do some extra digging and analysis to see how much more than normal might be flowing in there.

There may simply be some speculation based on people who want to make money in the unlikely case where Yellen doesn’t raise rates. Those who are expecting rates to rise as telegraphed previously have most likely already taken their positions, but people who are betting against it are more likely to wait close to the hour. If that is true, then one would expect a pattern that is slightly more significant than noise but not by much. But I would think that kind of thing would be done best at the near end of the yield curve. The far end has more duration, but is also less correlated to fed funds rates, so you’d be magnifying noise as well as signal.

The other possible thought is that there is an uptick in those who think raising rates will be negative for the economy (and by extension, stocks), and so the desire for long term fixed income might creep up.

Long term rates are affected by all sorts of things that I am not brushed up on: insurers, real estate, mortgages, long-term economic projections, and such. Ohai may well be right that it is unrelated or only indirectly related to expectations of a rate hike.

Historically has the market responded to the federal reserve rate like you suggest bchad? Seems strange people would be discounting cash flows at the Federal Reserve benchmark rate, as opposed to something like the 10 year treasury

Who is discounting cash flow at benchmark rate?

People usually discount cash flows at the swap rate (not treasury or fed funds), since this is easily traded in the market. I’m lost after the text wall, so sorry if I misunderstand some question…

I did make an assumption of a parallel shift in the yield curve, which in retrospect is a weak assumption. However, there are certainly some investors who think that the Fed will never be able to raise rates, whether for economic or political reasons, and so evidence that this actually can happen can make people update their views across the yield curve.

The front end and the back end are weakly correlated, but they are correlated. On the one hand, the correlation is weak but the duration is large. It is true that when moves are telegraphed in advance, it allows the yield curve to price things in and reduces the likelihood of parallel shifts.

I’d have to go back and check the data when I have the chance, but if rate cuts have the tendency to produce market increases, it seems plausible to assume that rate rises would have at least a dampening effect. I agree that this is an empirical question that I’m not current on (though perhaps I should be: I’m starting to get lazy).

I believe market performance is tied to the volatility of rate hikes, with larger/faster hikes leading to worse performance.