CMBs: More Questions Than Answers - Views from the North

Ben Reitzes:

Welcome to "Views from the North," a Canadian Rates and Macro podcast. This week I'm joined by Fred Nastos, BMO's head of Government Spread Trading, and Trina Reid, BMO's head of CMB Mortgage Aggregation Funding. This episode is titled "CMBs: More Questions than Answers."

I'm Ben Reitzes, and welcome to "Views from the North." Each episode, I will be joined by members of BMO's FICC, Sales, and Trading Desk, to bring you perspectives on the Canadian rates market and the macro economy. We strive to keep this show as interactive as possible by responding directly to questions submitted by our listeners and clients. We value your feedback, so please don't hesitate to reach out with any topics you'd like to hear about. I can be found on Bloomberg, or via email, at benjamin.reitzes@bmo.com. That's Benjamin dot R-E-I-T-Z-E-S at bmo.com. Your input is valued and greatly appreciated.

Trina and Fred, thank you for coming. Very interesting days, and I expect that you both will clear things up for me and everybody, so thank you for coming on.

Trina Reid:

No problem.

Fred Nastos:

Great to be here Ben.

Ben Reitzes:

We got the Federal budget yesterday. It's currently Wednesday, so we got the Federal budget on Tuesday. Outside of the mammoth spending that they always put in there, at least for the past few years, there was a surprise for markets. The Federal government told us they intend to undertake market consultations on the proposal to consolidate the Canada Mortgage Bonds within the government's regular borrowing program. Effectively, they'd roll up the Canada Mortgage Bond program into the government of Canada's regular borrowing. For all of the listeners out there that are not deep in markets, and especially not deep in the Canadian market, Fred, can you explain to us what the Canadian Mortgage Bond program is?

Fred Nastos:

The CMB program, Ben, is one of the funding tools that is available in the mortgage market for Canada. What occurs is, a portion of all the mortgages that are issued are collected, pooled together into mortgage-backed securities, or MBSs, and then some of those MBSs are purchased by the CMHC through a conduit called the Canada Housing Trust, with proceeds from a bond issuance that they do called the CMB program. CMBs trade on a secondary market, and there's price discovery. They trade with a spread back of Canada bonds. It varies depending on the term of the bonds, but it's around let's say 30 basis points. Investors buy the CMBs, the cash from the CMB issuance is then used to buy these mortgage backed securities, and then that money flows through the system, effectively helping finance a lot of mortgages in the Canadian market.

Ben Reitzes:

Okay, let me stop you there for a second. Rewind a little bit. This program started in 2001. Why was it put into place? I think it's important to start there.

Fred Nastos:

Sure. In 2001, MBSs existed, they traded, but they were very illiquid products. They traded with a spread back of Canadas, the MBSs themselves were guaranteed by the Federal government, by the CMHC, but they were very illiquid, and a lot of investors shunned them. They just weren't that interested in participating in them. When you own an MBS, you have exposure to prepayments from mortgages. There's a lot of administrative work just around managing a position in those securities. There was a wonderful idea to create the Canada Housing Trust, that would basically take care of a lot of that administrative work, and it would issue a bullet bond, a plain old vanilla bond, and use the proceeds of that to buy a collection of all these different MBSs, creating one liquid asset for investors. Then, the Canada Housing Trust manages the prepayment risks.

Trina Reid:

I think the offshoot of that is that the Canada Housing Trust takes no risk. Canada Housing Trust then has, again, highly rated institutions face them in a total return swap, with the MBS flows and the bullet bond flows.

Fred Nastos:

They manage that risk, right?

Trina Reid:

The banks manage, we manage that risk on our side, for CHT. We enter into a swap to guarantee the CMB coupon, and the CMB principle, and we receive the MBS principle and the MBS interests from CHT. CHT is just facilitating the cash flows between the CMB and the MBS, mismatch in duration.

Ben Reitzes:

Okay. That last part, facilitating cash flows between the two sides, is effectively what these bonds are doing, the Canada Housing Trust's role in all of this. If they're just a middleman, where is the issue here for the government? What they've highlighted, and this was in the budget, "Despite carrying the same credit rating, CMBs are a more costly form of borrowing compared to regular Government of Canada Bonds." Fred mentioned earlier, they trade at a 30 basis point spread, or depending on the term, just call it 30, to be nice and round. Should they be flat to Canadas? Why do they trade with the spread in the first place, if it has the same guarantee? I think this is one area where there is a fair amount of confusion, and just lack of knowledge, and I'd be guilty of that too, until probably today. As to why CMBs trade back of Canadas, can you please enlighten our listeners as to why there is a spread there over Canadas?

Fred Nastos:

That's a very common question, and every time we meet new investors to Canada, who are interested in Canada Mortgage Bonds, they ask, "Where does the spread come from? Why is there a spread if it's government guaranteed?" We have to keep in mind that the CMB program is effectively a clearing rate for all these MBSs. There's a liquidity risk that clients take on when they buy these bonds. They're a different form of collateral for a lot of the participants in the marketplace. They can have different value in repo, for example. Those different factors come together to add a spread that, investors need to be compensated. If there was no spread, investors wouldn't buy them. They would just buy Government of Canada Bonds. That spread is varied. Before the financial crisis, the CMBs would trade in maybe an average spread of around 15 back of Canadas. Post financial crisis, they've widened out to high twenties, low thirties, that area.

Trina Reid:

I think there's just housing market attached to that spread. There's a housing market premium there. This is perceived exposure to the housing market, which, we know all of these mortgages are all default insured by CMHC, Sagen, or Canada Guarantee. There is no risk to the investors. They have the stamp of CMHC guarantee on the security, both MBS as well as the CMB. I think that premium, on top of the reasons Fred mentioned, is also just related to the housing market, news headlines and stuff.

Ben Reitzes:

Maybe then the move here by the government is warranted. The reality is, there shouldn't be housing market attached to them. There's no real risk from a housing market perspective.

Trina Reid:

Well, it is a mortgage.

Ben Reitzes:

I understand that part. There's no default risk.

Trina Reid:

Correct, no default risk, but if you're buying the CMB, you are not really taking exposure to the underlying mortgage because you're not exposed to the prepayment risk, and you're not exposed to all of the underlying factors of the mortgage. If the market could have taken MBS back in the time, and was okay with amortizing securities, the market could have evolved very effectively then, but it didn't, and I think the CMB was a good substitution. It provided this ability for a lot of the smaller lenders to access liquidity on quarterly cadence, that bought all kinds of pools, no matter who issued them, and what size they were, and what pool type. There's also various pool types that CHT would buy, that the market would never see. I think there's lots of reasons it's proved very helpful for smaller lenders that are involved in the insured mortgage market.

Fred Nastos:

To add on to what you're saying there, the 30 basis point spread that we're referring to is basically a drop in the spread of the MBS. It's an upgrade, in a sense. The MBSs are trading wider than that.

Ben Reitzes:

Where does MBS trade now?

Trina Reid:

Again, the weighted average life of a four year MBS I think is probably high forties over the four year. The CMB is 30 over the five year.

Ben Reitzes:

So it's 10 to 20 tighter?

Trina Reid:

From an issuer's perspective, we buy mortgages, and we pool them. There's fees attached that are not just the spread you see in the trading world. There's fees attached to creating securities, and depending on how you fund them. My cost of funds is not just the spread on the product, as it is for other lenders. Sometimes, the cost of funds in MBS is cheaper, as it was through COVID. A lot of folks would fund by selling MBS directly to the market, and they didn't need the CMB, but curve inversion, et cetera, lots of things that have happened have made the CMB the cheapest cost of funds. It ebbs and flows, but the liquidity component of the CMB is the key piece. It's always there to buy no matter what, through any disruption in the market.

It has always been there to buy MBS from small and big lenders, and provide that liquidity. As Fred mentioned, that spread over CMBs is purely a pass through to the borrower based on the cost of funds that they will get for that insured mortgage. If it's 30 basis points over CMBs, me as an MBS seller would bear that cost, that 30, and if it's 50, CMHC wouldn't really care, because they're not paying that cost. It is me as an issuer that is paying that cost. I would be then be forced to pass that cost on through to the borrowers.

Ben Reitzes:

That last point is key there. The government's not actually paying these wider spreads.

Trina Reid:

The government does not pay these wider spreads. The government does not manage the risk on the underlying mortgages. In the current scenario, that is done by all the swap counterparties.

Ben Reitzes:

If you were to look at the government's financial statements, there is no line item for Canada Housing Trust interest. They actually just get a dividend from CMHC. The money flows the other way.

Trina Reid:

That would be the guaranteed fees that I pay to them.

Ben Reitzes:

The money flows to the government, not away from it. In the budget, the budget noting the higher borrowing cost, they really don't even bear that.

Trina Reid:

Are they talking about the higher borrowing costs for the government, or the borrower? All of this leads to higher borrowing costs, potentially, for the borrower.

Fred Nastos:

You have to assume they're referring to their borrowing costs, because they talk about-

Trina Reid:

GOC issuance.

Fred Nastos:

The GOC issuance, and they talk about, reducing the debt charges gives them a chance to reinvest the savings.

Ben Reitzes:

Okay, this is the next question I have for both of you. How do they capture that spread? What are they going to do? How will they structure the mortgage market, the Canadian mortgage market, to ensure what they say is stable access to mortgage financing, and maintain the status quo to some extent, at least for the end user, but change the CMB program?

Fred Nastos:

I feel this is the first point that needs to be clarified, exactly how will the program be implemented? There's various different ways you can envision this. One is that the CHT is dissolved, and as the government issues more Government of Canada Bonds, they take the proceeds of that issuance, and they just purchase MBSs, effectively from your desk, right?

Trina Reid:

Yes.

Fred Nastos:

Another way that they could implement this is, they could keep the CHT, still manage it, and as they issue more Government Canada Bonds, use the process of that just to fund the CHT, very much like the equivalent to a CMB issuance.

Trina Reid:

They could buy the MBS, or they could buy the CMB.

Fred Nastos:

Yeah.

Trina Reid:

They could issue the CMB, and be the sole investor in the CMB, and get the returns, that 30 basis points, on that side.

Fred Nastos:

That's right, yeah.

Ben Reitzes:

Would they actually make money? Is there actually benefit at the end of the day for them?

Trina Reid:

The reason they have a dividend that they pay out is because they get guaranteed fees from the folks that pool. We'll still pay that. They could take that dividend they use now, and use that for affordable housing, or whatever initiatives I think that they have on their list of things to do. I think potentially, by taking the investors out of the equation, it may give them some more flexibility on achieving the goals that they want in affordability, potentially. Again, speculation around how they plan to roll this out, but through their ESG goals and affordability goals, they could potentially not have to worry about an investor at the end of that whole program. They could just incent me, the issuer, by saying... if their goal is to generate more affordable product, they're going to incent me in a way to say, "Your guaranteed fees are half price if you go out and generate a bunch of affordability product."

It allows them to influence the market directly through the pooler and the lender, whereas now, they're a bit focused on the investor base, and satisfying the investor base in the CMB market.

Ben Reitzes:

There is a potential benefit for them from this, beyond just the money. They can change the structure of the market.

Trina Reid:

I think the money stays the same. Again, the guaranteed fees that are paid, assuming they still get us to pay them... if they're holding the MBS, I'm not sure why they would need a guaranteed fee paid by them to pay themselves, unless they're fundamentally changing that structure. If there's guaranteed fees still to be paid, then they could use those guaranteed fees to do whatever it is, and I suspect they're not going to give us GOC cost of funds. It's going to be GOCs plus some spread, potentially. There will be some margin there for them to then invest in the initiatives they would like to execute.

Ben Reitzes:

Okay. Where could this go wrong? Why might this be an issue? There will be consultations over the next few months, surely. Finance will go out and talk to all the banks, talk to all of the investors, or many investors at least, and canvas, to see what the market's thoughts are, and participant's thoughts are. What will be the objections to ending the program?

Trina Reid:

Why don't I give the mortgage originator point of view, and then Fred can jump in and give the whole trading point of view from investors? Mortgage originators, again, like I said, a lot of my clients would be smaller mortgage originators that have depended on the liquidity, so I think there's going to be a huge focus around that liquidity still being there in some way, shape or form, like the CMB was for them. If it isn't, and if that cost is going to be more variable, and not as consistent as the CMB, I think it will definitely have those folks rethinking how they're going to play in the insured mortgage market. Again, the insured mortgage market is generally just one piece of their funding needs, so if this becomes less stable and more expensive, they may just have to pivot, and fund their insured mortgages the same way they fund their uninsured mortgages, which could mean more pressure on the GIC market.

They're going to have to find those funds somewhere else, because this won't exist for them. I think, from that perspective, on the single family at least, that will be one thing that they will look at. I think in the multi-family space, which is a little bit of a different animal, again, the multi-family space is almost entirely consumed by the CMB, and that liquidity is very important for that market. The hedging off the back of the multi-family business is also, again, against the CMB. For mortgage originators, I think they need to see that. They need to know what that spread will be so they can originate a pipeline, and not have that variability, and need to warehouse it much longer than they thought because there's market disruptions, et cetera. I think that will be the main focus for the mortgage originators.

Ben Reitzes:

It sounds like there's some risk that mortgage rates could move higher if that liquidity is not there when they need it.

Trina Reid:

Absolutely, yes.

Ben Reitzes:

All right.

Fred Nastos:

Do you think Trina that the Federal government would have to guarantee a spread throughout?

Trina Reid:

I think they could do a fee-based thing, based on the term, again, through the guaranteed fees potentially. A very similar type of scenario, where folks can go out and originate mortgages, knowing what the clearing price is that they can get off those mortgages.

Ben Reitzes:

Should that vary with credit conditions?

Trina Reid:

That's a good question.

Ben Reitzes:

You don't have to answer it.

Trina Reid:

They potentially could reset it a couple of times a year. Again, mortgage pipelines are generally 30, 60, 90 days, so I think getting that answer in the middle of-

Ben Reitzes:

Like a pre-approval, you can just have it locked in for the next 90 days.

Trina Reid:

Exactly.

Ben Reitzes:

And tomorrow, change it for the next 90 days, and so on and so forth. Okay. Fred, what about the investor side of things? Where can all of this go wrong?

Fred Nastos:

I don't know if it'll go wrong, it's just more-

Ben Reitzes:

Where could there be problems?

Fred Nastos:

First, the Federal government will actually need to raise a lot more money. The CMB program is about $260 billion if they do about $40 billion a year of CMB issuance, and that will be issuance that the government will have to take on at the Federal level. It's unclear to me that issuing another $40 billion every year in perpetuity will come at the same levels as Government of Canada Bonds do right now. Government of Canada Bonds are very expensive as far as global bonds go. Our government gets a great borrowing rate for their programs. One measure that we use is where bonds traded versus OAS. The OAS market defines what the market expectations are for short term rates over long periods of time, and Government of Canada bonds typically trade around OAS flat.

Government of Canada bonds trade at OAS flat, a U.S. Treasury trades around OAS plus 20. If suddenly the extra $40 billion a year of issuance, or it could be more, as you were explaining.

Trina Reid:

It could be, yes.

Fred Nastos:

If that cheapens up Government of Canada bonds to be, just pick a round number, OAS plus five, then suddenly, the extra carry that they're making off $40 billion of CMBs that they're buying effectively goes away, because they have to reprice their entire stock of issuance at a wider spread.

Ben Reitzes:

That stock of issuance is over a trillion dollars, just to toss it out there.

Fred Nastos:

That's right, yeah. The CMB market represents about 25% of Federal issuance.

Trina Reid:

That's significant. I didn't realize it would be that much.

Fred Nastos:

We'd have a government where 20%, so if you have $250 billion of 1.3, 20% of their debt would be to finance mortgages.

Trina Reid:

Is that any other country doing this?

Ben Reitzes:

I don't know. I can't answer that. More of the question is, is that what we should be doing? Is that what the government's job is?

Trina Reid:

I can say, the closest thing to my program, from what I've read, in the U.S. is Ginnie Mae, and they have the underlying mortgages insured, as well as the security pool, but they certainly don't have the bullet solution for investors. They just sell the pass through, through MBS. The government isn't funding that. They're guaranteeing it, but they're not funding it. It does become a different conversation, I think, when you have the government actually buying the mortgages.

Ben Reitzes:

Definitely, I think that's where my head starts to spin. I can't imagine the government owning my mortgage, it doesn't make much sense to me. I don't think I want my government exposed to the housing market. Then, the economy's double exposed. One difference though the U.S. has 30-year mortgages, and we do not.

Trina Reid:

They have various terms mortgages, yes.

Ben Reitzes:

They have the capability.

Trina Reid:

They also have a lot more prepayment risk than we do here. The risk to be managed is very well known by investors, and they're very comfortable with it. In Canada because there's actual prepayment penalties on the underlying mortgage, it deters a lot of folks from prepaying their mortgage. There is not a lot of fluctuation. There is definitely fluctuation when rates move lower, there are some prepayments, but it's certainly not what we're looking at in the U.S. That is definitely a different risk management criteria.

Ben Reitzes:

Why are MBS so hard to trade here? Why is the liquidity so challenging?

Trina Reid:

The MBS also pays indemnity fees when rates do go lower, you get made whole. It could be said that the reason MBS is so difficult in Canada is because the CMB exists, and the second part of that-

Ben Reitzes:

Isn't it the other way around, though? Doesn't the CMB exists because the MBS were challenging at the time?

Trina Reid:

There is a point where you will limit the growth because you're not reaching that market, the other piece of that is, pre-2011, accounting rules stated that when you securitized and sold those MBS pools, you got that off your balance sheet. Because that is no longer the case, the banks don't have the benefit in selling those MBS. They don't sell those MBS to the market, so they don't get balance sheet relief from selling the MBS to the market. That's another key difference between us and the U.S. market as well.

Ben Reitzes:

Okay.

Trina Reid:

The other part of what else is impacted besides Fred's world, in the spreads and trading, and my world in mortgage aggregation, there's also the seller swap component of the existing CMBs that are in the market. Those are fairly significantly-sized books where, again, like I mentioned earlier, the CHT trades that are done with the dealers and the banks, to manage the risk off the amortizing mortgages and the bullet CMB, are them in themselves big buyers of MBS. If that goes away, there no longer is a need for that replacement asset component, for MBS purchases inside of that program as well.

Ben Reitzes:

Another aspect that was brought up today at various points, CMBs are a HQLA asset in Canada, and if you eliminate $260 billion of that asset, what replaces it?

Fred Nastos:

Yeah. That, from the investor side of course, is a key question. Before we even go there, I think there's another question of, what happens to the stock of CMBs that are existing in the marketplace right now? How will they trade? We've seen spreads on CMBs tighten seven to eight basis points since the announcement. There's definitely a scarcity premium attached to them, but there is a risk that, after a few years, they become what we call orphan bonds. They don't have a lot of focus, and don't trade very readily. To your point though, they make up a large part of the HQLA assets that are owned by Canadian banks, and those assets will need to be replaced. Every Canadian bank has a portfolio of high quality liquid assets that they can use if they ever need to raise liquidity, that they can sell, and CMBs make a larger part of it. That will need to be replaced with alternatives.

Ben Reitzes:

Which are?

Fred Nastos:

The main alternative would be provincials.

Ben Reitzes:

Okay.

Fred Nastos:

This actually has a slight positive to provincial issuers. It should help their spreads, especially in the short end, which is where a lot of these HQLA assets sit. The five-year part of the curve isn't necessarily a part they issue very regularly, but if suddenly the spread of five year provincials to Canada is tightened enough to the point where it's almost equivalent to CMBs, that might entice the provinces to issue a lot more in that space.

Ben Reitzes:

I guess that opens the door then, if the CMB program is taken away, for the provinces to issue a lot more in the five-year sector, and we can see them really pick things up there, especially if spreads tighten decently. Let's wrap up on one last point. Trina, can you tell us how big the mortgage market is, and how important CMBs are relative to the housing market as a whole?

Trina Reid:

If we look at just the prime market in Canada, I think it's fair to say that the size of the whole market would be about $1.8 trillion residential mortgages. The CMB and the NHA/MBS only fund the insured mortgage product, which is a portion of that overall $1.8, and that number in total rolls in at about $450 billion, of insured mortgages that are funded via MBS, via the CMB, and again, various other ways to fund. The program itself, again, the CMB, is funding only the insured component, through MBS.

Ben Reitzes:

It's a big part of the market, but the market is far larger than just the insured part of things.

Trina Reid:

Overall, it's been shrinking since 2016, and the insurance rules have changed. I think there was a time when that was closer to 60%, that insured number was 60% of the market. It's been trending downwards since 2016, and the single family specifically has been trending lower, whereas the multi-family insured component has been growing.

Ben Reitzes:

Okay. Thank you both for coming in and chatting with me today. Clearly, there are still a lot of questions that need to be answered here. I'm sure we'll be consulting with the government at some point in the next few months, at least you guys will be, and we'll see where things pan out. Stay tuned, still lots of questions, we'll see if we get answers, and hopefully, this resolves in a market-friendly, borrower-friendly, everybody friendly way, because we could sure use that.

Trina Reid:

Thanks Ben.

Fred Nastos:

Thank you Ben.

Ben Reitzes:

Thanks for listening to "Views from the North," a Canadian rates and macro podcast. I hope you'll join me again for another episode.

Speaker 5:

The views expressed here are those of the participants, and not those of BMO Capital Markets, its affiliates, or subsidiaries. For a full legal disclosure, visit bmocm.com/macrohorizons/legal.