Serving the Quantitative Finance Community

 
User avatar
Edgey
Topic Author
Posts: 14
Joined: March 23rd, 2005, 11:01 am

[Request] Fixed vs Floating irates - Existence of a risk premium

July 11th, 2016, 10:03 am

So I got into an office argument about whether or not there is a risk premium for issuing debt as a floating rate risk rather than fixed (ignoring credit spread effects).  What I'm looking for is an empirical paper that shows that a premium doesn't (or does) exist.  Can anyone point me towards a source.  

Cheers
 
User avatar
list1
Posts: 827
Joined: July 22nd, 2015, 2:12 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 11th, 2016, 4:23 pm

[Request] Fixed vs Floating irates - Existence of a risk premium viewtopic.php?p=787139#p787139by Edgey » Mon Jul 11, 2016 10:03 am
So I got into an office argument about whether or not there is a risk premium for issuing debt as a floating rate risk rather than fixed (ignoring credit spread effects).  What I'm looking for is an empirical paper that shows that a premium doesn't (or does) exist.  Can anyone point me towards a source.  

Cheers


If we ignore credit risk then debt instrument is similar risk free debt rate. Then in such setting what does it mean risk premium?
 
User avatar
Edgey
Topic Author
Posts: 14
Joined: March 23rd, 2005, 11:01 am

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 11th, 2016, 4:56 pm

It seems this new formatting has some teething trouble.  Anyway.

An example.  
I can issue fixed at 2% for 10 years
or
I can issue floating at 0.5% now and take a gamble on future rates.  

No arbitrage/financial theory tells me that I should be indifferent.  On average the floating rate will follow the forward curve and I'll end up paying coupons of (say) 2% in 5 years time and 3.5% in 10 years.  

BUT

Does this really happen in practice?  Can issuers of floating debt actually borrow cheaper on average over the next 10 years because they are exposing themselves to floating rate risk?  Recent history would suggest floating is better implying that there is a floating risk premium. I argue that this is mainly because yields have been declining, but I can't prove it is only because yields have been declining.  What empirical evidence can I cite to say that there is or isn't a systematic risk premium for issuing floating debt?
 
User avatar
list1
Posts: 827
Joined: July 22nd, 2015, 2:12 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 11th, 2016, 9:07 pm

It seems this new formatting has some teething trouble.  Anyway.

An example.  
I can issue fixed at 2% for 10 years
or
I can issue floating at 0.5% now and take a gamble on future rates.  

No arbitrage/financial theory tells me that I should be indifferent.  On average the floating rate will follow the forward curve and I'll end up paying coupons of (say) 2% in 5 years time and 3.5% in 10 years.  

BUT

Does this really happen in practice?  Can issuers of floating debt actually borrow cheaper on average over the next 10 years because they are exposing themselves to floating rate risk?  Recent history would suggest floating is better implying that there is a floating risk premium. I argue that this is mainly because yields have been declining, but I can't prove it is only because yields have been declining.  What empirical evidence can I cite to say that there is or isn't a systematic risk premium for issuing floating debt?
It seems that that you have something in your mind that does not represented in the message.
 Do you ignore risk free rate? 
"I can issue fixed at 2% for 10 years" // Does it mean that :
1. Company A can issue 10 bond with 2% fixed interest
"I can issue floating at 0.5% now and take a gamble on future rates  " 
2. It sounds somewhat indefinably.
 
User avatar
Paul
Posts: 6604
Joined: July 20th, 2001, 3:28 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 11th, 2016, 9:22 pm

An example.  
I can issue fixed at 2% for 10 years
or
I can issue floating at 0.5% now and take a gamble on future rates.  

No arbitrage/financial theory tells me that I should be indifferent.  On average the floating rate will follow the forward curve and I'll end up paying coupons of (say) 2% in 5 years time and 3.5% in 10 years. 
That's wrong. But it depends what you mean by "on average."
In the risk-neutral world this is correct.
But not in the real world.
Take the classical case of an upward-sloping forward curve. Think of this as depending on what is happening in the risk-neutral world. And yes the forward curve is where the short rate is expected (i.e. on average) to go in this world. But that's the funny old risk-neutral place.
Here on planet earth the short rate will typically just tootle along sideways, ignoring the forward curve. Now imagine the concept of an average real short rate. Plot this on the same graph as the forward curve. It will be much shallower than the upward-sloping forward curve. What you've now got is this average floating rate you mentioned. But in the real world. You've got to pay attention to what's real and what's risk neutral.
The difference between the two curves is...drum roll...due to the market price of interest rate risk.
And there's a brilliant paper on this precise topic. By whom you ask. Me, of course! With Riaz Ahmad. It used to be in the articles section of the site. I'll prioritize its reposting.
P
 
User avatar
Edgey
Topic Author
Posts: 14
Joined: March 23rd, 2005, 11:01 am

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 12th, 2016, 2:36 pm

List1 - Paul restates my question far better than I could explain.  Please let me know if I can clarify further.  


Paul - Found a copy.  Thanks! Paper link
A very interesting read.  Just what I was looking for.  I'll add some comments if I may:-

The sample period covered in the paper is one of decreasing short and long interest rates, with a mainly upward sloping curve i.e.
Image
I'm guessing this model would have performed well in the past 10 years, but I wonder if it would perform well if rates were to rise in the future?  
What if the parameters were calculated using the data from 1965 - 1985?  Would there still be an average Lamda = -1.2?  

I suspect a literature review was sacrificed for reasons of space.  When you were writing the paper, did you come across any other interesting research published on the real world evolution of interest rates?

Thanks again for your time and help.
 
User avatar
Paul
Posts: 6604
Joined: July 20th, 2001, 3:28 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 12th, 2016, 3:23 pm

That link is to a website by someone, "Planchet," who doesn't respect intellectual property! But it is the correct paper.

It doesn't matter whether the forward curve slopes up or down. If it is downward sloping it means that real rates are expected to fall even faster! That's assuming that the market price of risk has the right sign. But as you see from the paper the risk parameter sometimes has the wrong sign, meaning people are irrationally paying to take risk.

There is always the assumption that the future will be like the past statistically. So the current lengthy period of v low rates is a problem. You can do the analysis again and will get something v interesting, short rates with two stable points, because of the bimodal distribution.

There is bound to be something in the literature but probably not much. To appreciate this you have to come at it from the right angle. Your typical sell-side quant never cares about the real world, and the buy-side person doesn't understand quant finance! We first did this at Nomura, on an arb desk where we were looking for stat arb opps, in the early 1990s.

P
 
User avatar
list1
Posts: 827
Joined: July 22nd, 2015, 2:12 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 12th, 2016, 6:15 pm

Can we refine the notion interest rate risk. In what units it is measured. I understand that we have interest rate dynamics equation and its solution but than the interest rate risk is still undefined though we talk about hedging and its price.
 
User avatar
bearish
Posts: 5188
Joined: February 3rd, 2011, 2:19 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 12th, 2016, 6:16 pm

As somebody who has spent a few years on both the sell and buy sides, I humbly take some exception to both characterizations (although I don't deny they may be true statements about the medians). We were kicking around estimates of the USD interest rate risk premium at Merrill 20+ years ago, settling in on around 10-15 bps per year of duration. In my most recent buy side experience it was a frequent topic of discussion, and we had at least two relatively independent approaches to the estimation. Oddly, the result hasn't changed all that much over the years, although duration isn't quite the right metric to use for this purpose.
 
User avatar
Paul
Posts: 6604
Joined: July 20th, 2001, 3:28 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 12th, 2016, 7:20 pm

There's always an exception!

But you get my point that it helps to have seen both sides. You also see big differences in the level and style of maths used by the two sides.

P
 
User avatar
list1
Posts: 827
Joined: July 22nd, 2015, 2:12 pm

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

July 13th, 2016, 3:53 am

Can we refine the notion interest rate risk. In what units it is measured. I understand that we have interest rate dynamics equation and its solution but than the interest rate risk is still undefined though we talk about hedging and its price.
I try to explain my confusion. Let us look at the paper. It explains market risk and its pricing as following.
"Unlike equities where a model for the ‘underlying’ leads to the Black−Scholes equation, fixed income has a twist. Because the spot rate is not traded it is not possible to eliminate interest-rate risk by dynamic hedging. "   
/// a) In theory one can consider that spot rate is a traded asset by assign it a notional principal.
/// b) in equity derivatives we hedge option by a portion of underlings or a share of underlying we hedge by a portion of options. Whether the statement that "Because the spot rate is not traded it is not possible to eliminate interest-rate risk by dynamic hedging" implies the use of derivative contract which has the spot ir u ( r ) by its underlying. 
"Contrast this with equity derivatives for which it is theoretically possible to eliminate market risk by delta hedging. If we cannot eliminate risk then we must know how to price it. "
/// In BS theory which admits a risk free portfolio the price of the market risk is a portion of the short stocks in the risk free portfolio.
"This amounts to modelling how much extra expected return is required for a ‘unit’ amount of interest-rate risk. "
/// This statement sounds a little bit indefinably. Is it possible to express it by a formula.  

"Once this is specified then we use this same ‘market price of risk’ to price all fixed-income contracts in a consistent way. See Wilmott (2006) for details. We will denote this market price of interest-rate risk by λ"
 
User avatar
freddiemac
Posts: 7
Joined: July 17th, 2006, 8:29 am

Re: [Request] Fixed vs Floating irates - Existence of a risk premium

August 3rd, 2016, 8:48 am

Why is there a risk premium on the floating rate necessarily and not potentially on the fixed rate?

I know the fixed rate is fixed so it is known with certainty but does not the risk concept arise when you compare both sides of the balance sheet? So for a household a fixed rate mortgage is not necessarily less risky than a floating one as their income may be related to rates (eg in a recession where the household may lose their job mortgage payment would fall with floating rates creating a hedge).  Similarly, cyclical companies could potentially lower their risk be funding themselves with a portion of floating debt. Of course, the opposite may also be true but I am not sure I understand really why floating rate would attract a risk premium...

Apologies for jumping in on this very interesting thread with a very basic question. I know that in your article you find different signs of the risk premium so that could potentially be seen as consistent with my question.