Using Futures Contracts To Hedge ARMs
Investor · Round Rock, TX · Member since 2010 · 8k+ posts · 4k+ votes
Has anyone out there in creative land used futures contracts to hedge interest rate risk on your ARM? I am wondering what futures to use for commercial loans with floaters past year 5 or so. Selling eurodollar futures to hedge LIBOR floaters may make sense.
Does anyone have experience implementing something like this? Is basis risk the main thing to account for?
I looked into this a couple of years ago and found it much harder in practice than it seems in theory. (and I am a futures trader by trade!)
This pretty complicated stuff and I never found anyone who had actually done it on an individual basis.
It all looks perfectly straightforward on paper. But then I would do a "paper trade" and use real examples and prices to put the hedge on and then create scenarios and prices in the future to simulate different outcomes. Then tally up how effective the hedge was. I kept coming out on the losing end in these scenarios and finally figured out why.
The problem occurs when you see that the futures market already has priced in certain expected rate increases over the coming years. (look at the Euro$ futures going out a few years).
So let's say you see that 1 yr LIBOR is 1% (just using round numbers for example. Future price would be 99.00) and you want to hedge for an adjustment that will occur in 2 years. You are thinking to yourself "I want to lock in that 1% rate". So to do this you have 2 basic choices. Sell contracts dated 2 years in the future, or sell the nearest contract and keep rolling that nearest contract each expiration (sell a new on every 3 months or so). Most people would sell the 2 year future. The problem is that the 2 year future is not priced at 1%. It is priced at 2% (price of 98.00) because the market is expecting that rate rise over 2 years. (again, example pricing)
So you are not actually locking in a 1% rate, you are locking in a 2% rate. This is fine as long as you understand that.
So then create a scenario to check if your hedge worked.
The easiest scenario is let's say rates stay exactly the same as they are now for the next 2 years. In a "hedge" situation, you would expect to make no gain or loss ignoring commission costs. However, in this case you would lose because that 2 year contract you bought that is now coming to expiration is now priced at the current rate of 1%. So you sold it 2 years ago at 98.00 and now you will have to close the hedge by buying it back at 99.00. You have lost $2500 per contract hedged but your rate on your mortgage is unchanged.
Hope that makes sense.
NOw, in other scenarios the market may have gone your way or it may also go the other way but in all cases you will have that 1% difference between what you thought you were locking in and what you actually locked in. So let's say the market does indeed do as expected and the current rate is 2% in 2 years. You will have no gain or loss on the hedge but your mortgage will adjust up a percent and you will pay more interest costs so you will be a net loser.
If you choose the other strategy of selling the closest month and rolling often, your outcome is a little better but the same scenario happens on a smaller scale each rollover plus you have greater transaction costs. So instead of losing $2500 on 1 contract, you might lose $200 each rollover for 2 years. Still $1600 plus 8 commissions.
The basic idea is that when you hedge you are not locking in "today's rates" as most people think of today's rates. You are locking in today's rate expectations for the future. And the market has already priced in the expectation of a rise. If the market rises MORE than is expected today, then you will benefit by that extra amount. But if the market rises less than the expected amount (as in the scenario where rates stay the same) then you will lose that amount.
If you are happy to lock in that expected rate because you think the expectation is too little and it will be even worse in reality, then go ahead, it works. Just realize exactly what you are locking in.
The moral is to make a timely hedge you must hedge not before the rate move happens, but before the EXPECTATION of the rate move is priced in to the futures. We are in a scenario now where the rate move has not actually happened but enough people think it will happen that an increase is already priced in.
Selling Eurodollar futures to hedge against LIBOR should work. What about TBill futures? There basis risk you mention could also be an opportunity for arbitrage between spot prices and futures prices that you should consider.
I found an interesting article about basis risk especially for consumer debt. It seems the article was written mostly for structured finance and ABS markets. But it does have a few interesting points about spreads between prime rate and LIBOR, for example.
http://www.securitization.net/pdf/Fitch/BasisRisk_15Sept06.pdf
I looked into this a couple of years ago and found it much harder in practice than it seems in theory. (and I am a futures trader by trade!)
This pretty complicated stuff and I never found anyone who had actually done it on an individual basis.
It all looks perfectly straightforward on paper. But then I would do a "paper trade" and use real examples and prices to put the hedge on and then create scenarios and prices in the future to simulate different outcomes. Then tally up how effective the hedge was. I kept coming out on the losing end in these scenarios and finally figured out why.
The problem occurs when you see that the futures market already has priced in certain expected rate increases over the coming years. (look at the Euro$ futures going out a few years).
So let's say you see that 1 yr LIBOR is 1% (just using round numbers for example. Future price would be 99.00) and you want to hedge for an adjustment that will occur in 2 years. You are thinking to yourself "I want to lock in that 1% rate". So to do this you have 2 basic choices. Sell contracts dated 2 years in the future, or sell the nearest contract and keep rolling that nearest contract each expiration (sell a new on every 3 months or so). Most people would sell the 2 year future. The problem is that the 2 year future is not priced at 1%. It is priced at 2% (price of 98.00) because the market is expecting that rate rise over 2 years. (again, example pricing)
So you are not actually locking in a 1% rate, you are locking in a 2% rate. This is fine as long as you understand that.
So then create a scenario to check if your hedge worked.
The easiest scenario is let's say rates stay exactly the same as they are now for the next 2 years. In a "hedge" situation, you would expect to make no gain or loss ignoring commission costs. However, in this case you would lose because that 2 year contract you bought that is now coming to expiration is now priced at the current rate of 1%. So you sold it 2 years ago at 98.00 and now you will have to close the hedge by buying it back at 99.00. You have lost $2500 per contract hedged but your rate on your mortgage is unchanged.
Hope that makes sense.
NOw, in other scenarios the market may have gone your way or it may also go the other way but in all cases you will have that 1% difference between what you thought you were locking in and what you actually locked in. So let's say the market does indeed do as expected and the current rate is 2% in 2 years. You will have no gain or loss on the hedge but your mortgage will adjust up a percent and you will pay more interest costs so you will be a net loser.
If you choose the other strategy of selling the closest month and rolling often, your outcome is a little better but the same scenario happens on a smaller scale each rollover plus you have greater transaction costs. So instead of losing $2500 on 1 contract, you might lose $200 each rollover for 2 years. Still $1600 plus 8 commissions.
The basic idea is that when you hedge you are not locking in "today's rates" as most people think of today's rates. You are locking in today's rate expectations for the future. And the market has already priced in the expectation of a rise. If the market rises MORE than is expected today, then you will benefit by that extra amount. But if the market rises less than the expected amount (as in the scenario where rates stay the same) then you will lose that amount.
If you are happy to lock in that expected rate because you think the expectation is too little and it will be even worse in reality, then go ahead, it works. Just realize exactly what you are locking in.
The moral is to make a timely hedge you must hedge not before the rate move happens, but before the EXPECTATION of the rate move is priced in to the futures. We are in a scenario now where the rate move has not actually happened but enough people think it will happen that an increase is already priced in.
Nice explanation Eric...thanks for sharing.
I will post on a technique I read about later today so that we can explore it some more. New comments are certainly welcome in the interim.
What about shorting a short-duration bond market ETF.
My guess is that anything traded on an exchange will suffer from the expectation issue Eric cited. Would ETFs somehow be immune from this David?
That is what the analysis I read did...In effect you are trying to "create" a fixed-rate product out of an ARM.
I will post on it more later when I have more time.
Sure you can do that comparison. The OP didn't provide any details for comparison. But it definitely matters in your analysis what the cost would be to just convert to a fixed. There are probably times when an ARM holder can't convert to a fixed for whatever reason so it is not always feasible to compare. My only point was to help accurately analyze the potential cost of a hedge.
I was considering this strategy for myself a couple of years ago right after the main part of the crisis when jumbo's were at a very high premium so I was looking at ways to hang on to my ARM until the market normalized.
As it turned out rates obviously came down nicely and the ARM worked until recently refi'ing.
I would say that in most instances today in the residential market, it would be cost effective to just convert to a fixed if you qualify rather than hedge. But the OP talked about commercial loans, so maybe that makes a difference.
I know I am delinquent on posting on this guys...sorry. This week has been crazy busy, but lucrative ;-)
I'll try to post details tomorrow.
I've thought about this, too. Great topic.
I think there are a ton of people out there who have floating-rate mortgages and who would love to lock in a rate, but they can't because their credit/income is bad or the house has fallen in value.
What people really need is an interest-rate swap, which Bryan is trying to create individually using futures contracts.
As has been pointed out, for so long as the rate outlook is "higher rates ahead," doing this could result in not being ahead financially when the dust settles. However, that's actually OK in times when the yield curve is normal, since if you're trying to lock in any rate -- be it through the mortgage market, the futures market, or wherever -- you should fully expect to pay a higher fixed rate than today's short-term rate.
I've always been a little surprised that one of the big investment houses hasn't offered this product to consumers. It would be SO easy for them to offer interest rate swaps and pick up spreads along the way. I think the problems are that (1) most consumers would not understand the product, (2) the public perception of the big, evil investment bank trying to seduce the public into any swap contract could be bad, and (3) psychologically it would be tough for a consumer to start writing checks on Day 1 of the transaction to the big, evil investment bank for the delta between the variable rate and the presumably higher fixed rate, never really knowing (or understanding?) that the big, evil investment bank would have to write checks back to the homeowner should variable rates shoot up.
And of course, the easiest solution -- for most people -- to get a fixed-rate loan is just to refinance into a fixed-rate product, and the banks love nothing more than refinances because they generate lots of juicy fee income, ergo they have little incentive to solve the problem by doing interest rate swaps.
I love this question, though. It's something I've thought about trying to implement. GMTA, Bryan!
Yeah...when I get off my keister (sp?) and post about the strategy it involves drawing on a LOC instead of writing checks. My guess would be that this is much less damaging psychologically than stroking a check when the bet runs the wrong way.
More to come...discussion in the interim is certainly appreciated.
Okay...finally got some time to write this post. So I wanted to point out that a hedged ARM also provides one the ability to qualify for a larger mortgage than the fixed-rate alternative because the start rate is lower. This is a nice benefit to using something like this instead of a fixed-rate mortgage too. You would also have the ability to prepay the mortgage in some scenarios without penalty, which could be used to increase cash flow from the project if there were no other good uses for cash at the time. This doesn't make a whole lot of sense right now, but it could if rates were much higher than they currently are.
Strategy:
You borrow a quantity of money on an ARM and hedge it by selling an interest rate futures contract equal to your loan. You could use treasury bonds with a coupon rate that equals the start rate for your ARM. You can roll these contracts every 30 months with minimal transaction costs.
The hedge is margined at about 4% of face value. Apparently many brokers will allow you to use interest-bearing securities for the hedge so you can earn interest on them pledged as margin. So the true cost is the difference between the rate you earn at and your required return on your equity. This is non-trivial, but is likely largely outweighed by the spread between the fixed-rate and the ARM start rate. Note that this is especially beneficial for large loans where the rate spread between FRMs and ARMs can add up to a lot of money.
If market rates rise appreciably the guy on the other end of the futures contract has agreed to buy for a fixed amount and you pocket the spread to cash in on the hedge. The broker can transfer an amount of money needed to you to cover the extra ARM payments.
If rates decrease the ARM payment would go down, your cash flow from the property would go up, and presumably the market value would rise too. Your interest rate futures contract would decrease in value and you would get a margin call. You would either need cash which would be very expensive or the ability to draw on a line of credit to cover the margin call. Presumably the increased value of your property could provide the value needed for the collateral for the line of credit or you could use some other real estate as collateral.
So that is the basic idea and I am sure it may suffer from some of the problems Eric described above. There is likely some basis risk and other things I am not accounting for.
Thoughts? It seems nice in theory and could potentially save people a ton of money when spreads between FRMs and start rates for ARMs are high.
Bump...any takers?
Again, you have described it well "in theory". But the problem lies in the performance of the hedge and the 30 month in the future EXPECTED rate that you have hedged at. That is what is going to ruin the idea.
Again, look at exact prices in todays markets and then extrapolate out into the future to see what the result of the hedge will be. I think you will find it doesn't accomplish what you hope.
FYI, you would not use Treasury Bond future to hedge. You would use either Tbill or Eurodollar futures depending on what your rate is based on. (Euro$ for Libor and Tbill for 11th district rate, etc)
To give you an idea. If you use Eurodollar Futures. The nearest contract is priced at 99.695 indicating a 1 month LIBOR of .305% but the 30 month out contract (June 2013) is priced at 97.735 so it projects the 1 month Libor to be 2.265% at that time. That is the rate you would be locking in.
So in your scenario. If you put the hedge on and short term rates go up 2% points. You will simply break even on your hedge. But your ARM mortgage would adjust up by 2% costing you that much extra in interest over the next year. Your hedge would not be effective. In most every scenario up or down, you would lose that 2% points. (in scenarios where rates go up more than the allowable adjustment on your ARM, you would win in the short term)
The other thing is this. I assume you are talking about using a 3/1 ARM since you picked the 30 month contract. #1, if this is the case you would have to use the 36 month contract which is priced another .5% higher. #2 This is kind of hard to explain...maybe you already knew this. Placing this hedge only protects you on the first rate adjustment, You would have another pending adjustment 1 year after that and another after that and so on. Each of those potential adjustments is a separate economic event and must be hedged separately. If you are trying to turn theoretically this into a FRM that means to lock in todays rate for those events too so you would have to place hedge NOW for those events too.
Again, with each set of hedges you run into the basis problem since the expected rate that is priced in to those is even higher.
This also means you have to have a good idea of how long you want to hold the property so you can decide how many hedges you have to place. If you want to hold 7 years, then you need to place 3 sets of hedges selling the 36 month, 48, and 60 month contracts. You don't roll them as the expire. Again, all this is fine if you are comfortable with the margin requirements, and the expected rate(s) you have locked in.....and if you are correct in you guess of how long you want to keep your property. Because if you are wrong on the 7 years and have to sell early, you have those "naked" hedges out there which could turn into large losers or winners, you just don't know. Of course, if you hold longer than 7 years, you would be unhedged for those extra rate adjustments.
I guess what I am saying is that your plan of turning an ARM into a fixed with hedges can't work the way you theorize in the real market because of basis and it opens up some unintended risks to boot. You are not turning your ARM to a FRM at today's rates. You are turning it into a sort of fixed ARM where you are locking your future rate adjustments based on the market's expectations today.
So again using actual market prices, you are essentially locking in and guaranteeing yourself a 2% rate increase on your first adjustment. That is a bad idea because 2% is probably the max adjustment you can have anyway, so you have actually locked in your worst case scenario.
It is all sort of hard to explain in words. Again, as I said, test out a sample and work out all the scenarios using real pricing and you will see the issues.
Okay...that makes sense to me. Thanks Eric.
I guess the lenders have accounted for this already in their loan products from what I am reading.
OK, I'll confess I didn't make it through all of the discussion, but I get the show stopper to be forward expectations of rising rates built into the instruments.
Someone back earlier in the thread mentioned interest rate swaps (or possibly a forward starting interest rate swap in the case of a 3/1 or 5/1 ARM) as the perfect instrument to accomplish this, but they're only available in the institutional space. Eric, Bryan, do you agree that an interest rate swap accomplishes this objective. I work for a large insurance company which utilizes many billions of notional of these to lenghten duration, convert fixed assets to floaters, etc. I'm not an expert on their usage, but I do know that we can use swaps to go out and lock in current fixed rates. We would recieve/pay a floating Libor-indexed rate which moves in tandem with our underlying floating asset/liability and receive/pay a fixed swap rate over the life of the term. If this is the case, I would have thought (maybe naively) that a futures contract would accomplish the same end.
I'll try to follow up with some of the deriv "experts" at my company on the topic. I also wondered, if you could in fact figure out a way to do the hedge, would it be cleaner and better fitting to get a loan that floats from the outset (rather than the 5/1 that commercial/portfolio lenders offer) to better match your derivative.
All a swap really is is a customized futures or options contract.
On the exchanges, the futures are standardized in size, etc. A swap is a transaction where you swap one risk for another risk, just like a future. But it is not standardized, so you can negotiate the exact size and details you want with the counterparty.
Yes, they are typically institutional.
The thing is you will run into the same issues with forward rate expectations. Pricing on all these products are very closely linked. SO pricing you get on Swaps trying to accomplish the same things will be priced very close to the futures contracts.
Guys, this can't be done. There is no way to lock in today's current rate into the future when the expected future rates are much higher. The market is much too sophisticated to allow that.
I do not think Bryan's goal is to lock in the ARM rate for a long period. His idea was whether he could get a lower fixed rate using a combination of an ARM + futures versus going in for a fixed rate mortgage.
The goal would really be to get either an equivalent or lower fixed rate than what is available in the market today while choosing an ARM.
Is this possible given what I described?
I'd like to bring this post back to life. I think it is very relevant and it's something I've been exploring for a while myself. Here's my situation:
I have about $1 million in ARM's (about 25 loans), all are indexed to 1 year US Treasury notes. Right now things are great because my rates are all around 3%. I'd like to lock that in.
The OP stated that he wanted to sell Eurodollar futures as a hedge, which is exactly what I've been looking at. My thought is to sell 4 contracts of the nearest contract and then continuously roll that over until I no longer need the hedge (most likely when I think rates have peaked). I understand that ED futures prices are expectational, however in a hedge one is not looking to capture actual value. One is really looking to capture RELATIVE value. In other words, I am just looking to capture price movements over the total period of my hedge. In my case, I would be looking to lock in rates over the next 5 years, so I'm looking to hedge against big movements in rates over that period. So even if ED rates have a built in 50 bp upward move that I will miss out on, it's not the end of the world because I would capture the balance of the upward move over the next 5 years.
Incidentally, I just checked and the Mar 2011 ED is at .3125%. 3 mo LIBOR is at .30% and the 1 year note is at .23%. This is about as close to parity as I've seen in a while for my particular hedge.
I understand that there other risk involved. In this trade the biggest one I can think of is parity loss between LIBOR and UST's.
Thoughts?
Ryan, could you walk us through how you came to select four as the number that you needed to use to hedge your loans?
It has been 20 years since I traded a Eurodollar contract, but my recollection -- and perhaps things have changed -- is that each contract is for $1 million, but because they are 90-day futures, each basis point (BP) is worth $25 per contract.
Is that still true?
If that is accurate, then I'm wondering how using these futures will really help you on a recurring basis. What I mean by that is this...
Presently each BP move on your collections of ARMs is going to add $100 in annual interest expense. So, if you short four Eurodollar futures contracts, and rates move up 1 BP, you gain $100. That offsets your added interest expense for the year; so far, so good.
But, what about next year? Assuming everything stays the same after that, your ARMs will still cost you $100 more each year, but you won't have any recurring profit from your futures positions.
How can you use futures on a recurring basis?
You know, with a million dollars in notional value, you just might find a bank willing to create an interest rate swap for you, assuming that you can post some collateral for the risk they calculate. That is really what you want here. I did this for clients in a past life, but of course it was all tied to the variable-credit product that we were offering. Maybe the bank wouldn't do a stand-alone swap. Can't hurt to ask, though.
I made a lengthy post on here about using a line of credit in conjunction with the hedge. Check that out and see if it may help.
If you end up doing this I would love to hear how it works out periodically. We have FRM for most of our product so it doesn't matter, but it would be nice to have the "best of both worlds" using a hedged ARM for larger purchases going forward.
@Deuce- great question. I will have to think about that. I'm headed out to show houses for the afternoon, so I will post back tonight or tomorrow morning. I've modeled this in Excel and it seems to work. But I am by no means a futures market guy...more of a real estate guy with enough knowledge of the financial markets to get myself in trouble. I have no doubt that I'm missing something.
@Bryan- I will read your post again and see if there isa better way.