Outlook: HSH's 2023 Mortgage & Housing Forecasts - Mid-year Review

Mortgage rate outlook

The past year was the most volatile period for mortgage rates in perhaps 40 years, as 30-year fixed-rate mortgages moved from a level just above historic lows to as high as they have been in more than two decades. The sharp and continued increases were both sizable and abrupt, upending the housing market, and as a result, most forecasts for mortgage rates were wildly off the mark, including ours.

But who knew that inflation would go from a long-smoldering ember to a roaring blaze in such a short time? Or that the Fed would have increased rates with a speed and ferocity rarely seen in its history?

Aside from roaring inflation driving long-term interest rates higher, one less-discussed influence that boosted rates was that the Fed -- the largest buyer of Mortgage-Backed Securities (MBS) for several years -- not only stepped out of the market for buying these instruments but also began steps to reduce the size its massive holdings of bonds, moving from Quantitative Easing to level support to Quantitative Tightening in a period of just a few months. We'll look more closely at this aspect of monetary policy in the Fed section below.

But the tumult of 2022 is behind us now, and while we'll never say never, the odds favor a much calmer market for interest rates for 2023.

To start with, while inflation of course remains a problem and a concern, it's no longer a surprise. Importantly, extreme increases in price pressures have leveled off, shifting from mostly goods-related inflation to more service-related. However, while the upturn in prices for goods may be fading amid softer consumer demand and supply-chain improvement, it's not clear if the same can be said for service-related inflation, which tends to be both stickier and less immediately responsive to changes in monetary policy.

Core PCE inflation -- the Fed's favorite measure -- so far has peaked at an annualized 5.3% rate back in February 2022, but hasn't moved much below that level or showed many signs of any sustained downturn. For example, core PCE managed to ease only to a 4.7% annualized by July before firming again to 5.2% by October. As we write this, core PCE is still running at a 5% annual rate, about two-and-a-half times the mark for which the Fed is shooting.

Important to all of this is that the inflation-producing disruption of the pandemic closure and re-opening is fading into the rearview mirror, as is the initial shock to supply chains from the war in Ukraine. There are of course still market impacts from both of these yet to play out, but as is the case with inflation and the Fed's significant response, the shock of these items has subsided, even if some of the impacts and implications from them will continue for some time.

What matters for mortgage rates in 2023 is whether the Fed gets inflation under control to the point where investor expectations shift from "how many more rate hikes?" and "how long will the Fed keep rates there?" to one of "the Fed is becoming more likely to cut rates soon." In this regard, it is how successful the 425 basis points of rate hikes in 2022 have been in cooling demand and quelling inflation, and how much progress toward the Fed's 2% core PCE goal has been made as the year progresses.

The bulk of Fed rate hikes are likely behind us, with just a few more yet to come. At the same time, it seems likely that inflation has peaked or is fairly close to that level now. But even if both of these things are true, the fact remains that short-term rates will still be lifted higher in 2023, inflation is likely to be slow to retreat (at least until later in the year), and a 2% level for inflation remains a good bit away from the level we see in December 2022.

When mortgage rates are rising and especially when they rise quickly, they tend to overshoot the level that conditions actually warrant. We've seen this time and again over the years, and the sharp runup to 20-odd year highs for mortgage rates by November has given way to a partial retracement, and perhaps one that has also gone a bit too far, so and rates may rebound a bit as 2023 kicks in.

While it's unlikely to be a smooth ride, we think that the average offered rate for a conforming 30-year fixed-rate mortgage as reported by Freddie Mac will likely manage to hold a 5.875% to 6.875% range for 2023. Should a recession form next year, there's a good chance that we'd see rates push though that bottom; conversely, if inflation doesn't behave as hoped, that top figure might not hold, either.

Mid-year review:
So far, so good. Compared to last year, mortgage rates have had a relatively calm first half of 2023, but there has been enough volatility to see the average conforming 30-year FRM wander in a 70 basis point (0.70%) range over the first 26 weeks of the year, but staying within the boundaries we expected. So far, 30-year FRMs have been as low as a comparatively cheap 6.09% and as high as a rather uncomfortable 6.79%, but our expected forecast range is still holding. Here's hoping rates spend more time toward the lower portion of our range over the next half of the year, but this may be difficult, given stubborn inflation and a Fed again pointing toward higher short-term rates before long.

The Fed / Monetary Policy

The Fed's abrupt shift from a cautious and "dovish" stance regarding monetary policy to a highly aggressive "hawkish" one was the driving force in financial markets in 2022. In addition to repeatedly raising interest rates, first mildly, then with increasing size and frequency, the central bank also not only wound down its pandemic-response QE program more quickly than expected but actually implemented a plan to begin to reverse it.

The Fed lifted the federal funds rate policy rate from a range of 0% - 0.25% that was in place as recently as mid-March of 2022 to one that now has a 4.5% top to close out the year. It's not that the Fed has never raised interest rates by 425 basis points before; rather, it's the speed with which they have done so which is remarkable, if not unprecedented.

The Fed apparently still thinks it has considerable work yet to do, unless it's simply "jawboning" the markets to keep them from getting ahead of where the Fed wants them to be. That happenstance occurred back in August, when Fed Chair Powell intimated that at some point the Fed would slow rate hikes or even pause them in order to assess the effects of so many rate hikes in such a short period of time. Markets took this to mean the Fed was nearing the end of this rate-hike cycle (it wasn't), and financial markets rallied and yields fell sharply for a time. This was a loosening of financial conditions the Fed did not and likely does not yet want to see happening.

But even if there is still work for the Fed, it's likely not all that much. The federal funds rate begins 2023 at a top rate of 4.5%, and the policy-setting committee's own projections suggest that this key short-term rate might rise as high as perhaps 5.25%. so this would only be an additional increase of three-quarters of a percentage point. Most likely, these projected increases will come in the early part of 2023, and possibly only in quarter point increments, although the cadence could be a half-point again at the end of January and a quarter point each in March and May.

Of course, what happens after that (if anything) depends on what happens to both inflation and especially the labor market. If inflation shows routine signs of lessening by May and reasonable progress toward the Fed's 2% goal -- say, core PCE declining by a percentage point in the first half of the year, to 4% or less -- this might be enough to put the Fed on hold. However, with "service" inflation still firm (or even firming as we write this) and more sticky, the labor market has to show more significant signs of becoming better balanced. This would be seen in the form of fewer job openings, a higher level of claims for unemployment benefits, an unemployment rate moving further away from near 50-year lows, greater labor force participation and especially wage increases settling back from around 5% per year to levels the Fed feels are more consistent with 2% inflation over time -- essentially, 2% plus whatever productivity growth can be added on top of it.

If all these variables are trending positive, the central bank may hold fire after May for the remainder of the year. If progress is grudging, additional policy moves may come.

But what of the Fed's program of reducing its balance sheet? In our weekly MarketTrends, we've detailed on a few occasions how the process of reducing its holdings of Treasury Bonds and especially Mortgage-Backed Securities are both missing the marks set for them (MBS most appreciably). For MBS, primary means of reduction -- refinancing by homeowners and trade-up homebuyers closing older loans the Fed currently holds (and in turn creating new ones for investors to absorb) -- has come to a virtual standstill. By our reckoning, the Fed's MBS holdings should have been reduced by $157.5 billion by the end of 2022; presently, it appears that just shy of $69 billion in runoff has actually occurred.

In 2022, the process of starting and then accelerating reductions of these holdings both contributed to lifting mortgage rates; that said, it appears that demand for MBS and supply of new bonds for investors to pick up is coming into far better balance. If the trend of diminishing home sales and subdued refinancing continues, there may soon not be enough MBS to satisfy routine investor needs. While this isn't yet the case, it is likely what the Fed is hoping for, since in order to meet its stated portfolio-reduction goals, mortgage rates either need to fall enough to kick refinancing somewhat higher and spark some trade-up home buying -- or more likely at this point, the Fed will need to begin outright sales of MBS holdings.

We think that by June, the Fed will lay groundwork for conducting sales of MBS and will start a small program of doing so late in the year, possibly September or thereafter, and likely only on an opportune basis -- that is, during times when new GSE-backed bonds for sale run below a given threshold, or when investors have bid up the price of bonds to a certain point, a reliable indicator that suggests demand for bonds is outstripping supply. Of course, this is pure speculation; we don't know the Fed's thinking in this matter any more than anyone else; they may simply use a fixed, programmatic approach, such as some kind of regular auction of these surplus bonds. Provided there's no market-disturbing wholesale sales of MBS and they aren't added into a market struggling to digest a given level of supply, there should only be a limited impact on mortgage rates overall (keeping them from declining a bit at times if any thing) although the announcement that sales may occur could cause a little bit of initial disruption.

Mid-year review:
Our expectations for monetary policy moves by the Fed have been fairly met. We did see the three hikes in the federal funds rate we expected (February, March, May) and there has been a pause, or at least a one-meeting (so far) skip by the Fed, as they refrained from raising rates in June.

Ostensibly, the decision to hold fire at least temporarily was in part dictated by the effects of bank failures and stresses earlier in the spring, a happenstance we didn't see nor expect. That said, the period of unchanged short-term rates may be short-lived; as we noted above, if inflation and labor markets were only making grudging progress toward the Fed's goals -- and that's all they seem to have made over the last six months -- that additional policy moves may come. According to the June Summary of Economic Projections, Fed members now expect another 1-2 quarter-point hikes in rates over the remainder of the year... unless, of course, sudden and significant additional progress on inflation or loosening in labor market should appear.

Also, despite mortgage holdings only falling off the Fed's balance sheet at a pace of less than half the desired level, we've also seen nothing regarding any selling of MBS holdings as yet. It's not clear if the spring's banking issues may have pushed such considerations to the back burner, or perhaps the debt-ceiling brouhaha was a factor. To be sure, it's also not clear if there would be enough appetite by investors to absorb such additional supply, and it's also true that the Fed would lose money by selling their low coupon holdings in the current market. Further, there's likely to be a cascade of new Treasury debt coming, and that alone could push yields higher, exacerbating losses and making sales more difficult. Absent a refinancing boom or significant surge in home sales, redemptions of MBS holdings will only occur at a snail's pace, so this still makes it likely that sales will need to be held at some point to meet balance-sheet reduction goals.

Mortgage Regulations

While the mid-term elections changed the political mix in Washington, we're only still in the middle of the current administration's term, so it's not clear if there are any new priorities for mortgages or housing that might be tackled. Much of the regulatory climate over the last couple of years has dealt with pandemic-related mortgage forbearance and such, but those issues and efforts are thankfully fading into the rearview mirror. However, that doesn't mean there aren't changes or potential changes afoot, but much of the regulatory attention at the moment seems to be happening largely outside of the mortgage space.A couple of items will affect consumers and mortgages to a greater degree in 2023, tho. In October 2022, Fannie Mae and Freddie Mac's regulator (the FHFA) changed the regulations governing the kinds of credit scores that the GSEs can use when granting mortgage loans. Fannie and Freddie have long adhered to a several-versions-old traditional credit scoring model from FICO, and the change allows them to start to incorporate far newer versions called FICO 10T and VantageScore 4.0. Rather than revealing a snapshot in time of a potential borrower's credit, these new versions used so-called "trended" credit data, looking at patterns of credit flow over a period of a couple of years. The new models may also include new account activity and histories for borrowers when available, such as rent, utilities, and internet or phone service payments. As well, the new models are said to also give extra weight to both credit utilization and missed payments.

While it is expected to take some time (possibly years) for the mortgage lending industry to incorporate these changes, the FHFA encouraged lenders to move as quickly as possible. With the lending climate soft, enterprising lenders may look to use these new scores as a competitive advantage, as it has been reckoned that 72,000 more borrowers will be eligible to obtain a mortgage using the new scoring model than the old. So, if you were a marginal borrower in 2022 and didn't make the cut, you just might make it over the bar in 2023.

We also mention one other interesting regulatory item that cropped up recently, but probably won't go anywhere in 2023. A few months ago the CFPB asked industry participants for feedback on a couple of unique proposals made largely in the name of expanding refinancing benefits for certain homeowners. Most specifically, the regulator sought feedback on "ways to facilitate refinances for consumers that would benefit from refinances, especially consumers with smaller loan balances; and also on ways to reduce risks for consumers that experience disruptions that could interfere with their ability to remain current on their mortgage payments" allowing for a kind of automatic forbearance for distressed borrowers.

The Bureau outlined ideas such as targeted and streamlined refinancing programs and processes for low-balance mortgage loans, pondered and discussed mortgage contracts that allow for loans to "automatically refinance" or even those that work as "one-way ARMs", where a borrower would get a mortgage loan whose rate could only go down, never up. While these are novel ideas, there are practical problems with trying to implement them, and the American Bankers Association and Consumer Bankers Association drafted a detailed response to the Bureau's request for information., outlining their takes on the benefits and market drawbacks of each.

Outide of these, and at least at the moment, it doesn't seem as though there will be much by way of a new regulatory push in the mortgage space for 2023, but a year is a long time.

Mid-year review:
There's not a lot of change on the regulatory front so far this year, but a consortium of mortgage trade groups has asked the FHFA for a longer timeline for the implementation of new scoring model usage for Fannie and Freddie. The FHFA would like to see full implementation by the end of 2025, but industry participants would like more time. Of course, this also allows future homebuyers more time to get their credit profiles in tip-top shape, too.

Fannie/Freddie/FHFA/FHA

The likelihood of any kind of reformation of Fannie Mae and Freddie Mac anytime soon is very slim or close to none, especially with a divided Congress and in the middle of an administration cycle. If anything, most likely is that there will be an even stronger push for Fannie and Freddie to promote and expand "affordable housing" goals, looking for ways to lower costs and improve access to mortgages for low- and moderate income buyers and those in underserved communities.

These include a new initiative revealed in December to allow Fannie to buy loans from borrowers who have no credit score and debt-to-income (DTI) ratios of up to 50%, using asset and cash-flow verification to assess the riskiness of a borrower. Other such efforts include eliminating up-front fees for first-time homebuyers earning up to 100% of an Area's Median Income (AMI) for first-time borrowers using HomeReady and Home Possible mortgages and for certain loans made through state home mortgage finance agencies.

Conversely, borrowers taking high-balance loans (aka "jumbo conforming" or "agency jumbo") have been seen increased fees, as have those borrowers who have conducted cash-out refinancings or financed second homes.

While more tweaks to programs and fees for low-mod borrowers may come in 2023, it seems unlikely that we'll see any sort of cost cuts for typical borrowers.

2023 could be the year we'll see some changes in costs for FHA borrowers, though. The FHA's self-insurance pool (MMIF) recently reported a capitalization ratio of 11.11%, a level more than four times the statutory requirement for the MMIF. FHA up-front and recurring mortgage insurance premiums have been in place for years now, and MI in the FHA program has not been cancelable for most FHA borrowers since back in 2013, so the only way for a homeowner to get out of paying the MIP is to refinance to a conventional loan.

A fair portion of the MMIF improvement has been the rapid appreciation in home prices, which lifts the value of the insured assets and lessens the severity of loss should a borrower default. Many FHA mortgages begin with just 3.5% down, and borrowers over the past few years have seen their equity positions deepen rapidly. However, we appear to be at the beginning of a cooling for home price increases, and may even see declines before long, which could see recent-vintage borrowers moving from a small starting equity position to an underwater situation while older loans aren't affected very much. While having a mortgage that exceeds the value of a home doesn't produce losses by itself, it can help produce them should a borrower encounter difficulty in making payments, the kind of situation that a recession would create, for example.

Given an uncertain housing climate, the FHA will probably proceed cautiously. That said, there's a reasonable chance that with the MMIF well over-capitalized -- and given lower-fee competition from Fannie Mae and Freddie Mac for low- and moderate-income income borrowers -- that some trim in MIP (whether up-front or recurring) may come this year. As part of their annual report to Congress, FHA ran a series of conditional "stress tests" against its insurance fund and even applying the most challenging conditions from 2007 still left the MMIF at about three times the required level of capitalization, so there seems to be enough space as to cut costs for some borrowers, at least a little.

We expect the efforts of Fannie, Freddie, FHFA and FHA to continue to be aimed at lowering costs for borrowers of lesser means in 2023.

Mid-year review:
Fannie and Freddie did make some controversial changes to risk-based loan-level pricing adjustments this year, ones that lowered fees for borrowers with lower credit scores or smaller down payments while increasing costs for borrowers with solid credit and more skin in the game. Some saw this as a kind of socialization of mortgage credit; others, simply the latest adjustment in loan-level costs to reflect a changing marketplace and a greater emphasis on the GSE's affordable housing mission.

It's fair to say that risk-based price adjustments should reflect the risk of the borrower, and that riskier borrowers should pay more as a result. That said, in looking at the construct of risk across the spectrum of originations and current market conditions, it's worth considering that the changes don't mean all that much -- that is, few homes available to buy, high home prices and high mortgage rates are likely seeing very few loans being made to borrowers at the margins of credit or LTV ratio anyway, and the changes don't really improve a buying or refinancing opportunity very much.

As an example, before the change, a borrower with a FICO 620 and a 5% down payment would have seen 3.25% in fees, which would have added perhaps as much as 0.875% to the loan's interest rate. Post change, the add-on will be 2.25%, adding about 0.625% to the interest rate. While this is a meaningful improvement, in the context of the current market, it means being offered an interest rate of perhaps 7.375% instead of 7.625%. It's a safe bet that very few borrowers are looking for loans or qualifying for financing at even the now-improved interest rate.

Conversely, adding some costs for borrowers with mid-to-upper tier credit isn't welcome, but really shouldn't change their decision to purchase or refinance, either. Pre-change, a borrower with a FICO 700 and 15% down faced a 1% fee; post-change, a 1.5% fee. This is an additional $500 outlay per each $100,000 borrowed, or would add about 0.375% to the rate, rather than the previous 0.25%. Again, not welcome, but not likely a decision-changing cost.

The GSEs have changed LLPA structures numerous times since their implementation, adding and dropping fees for things like "adverse market conditions" and other reasons, and fee additions, changes and even deletions chances happen all the time. Some are even announced but never make it to market; as an example, a new debt-to-income-based LLPA that was to be charged to borrowers with DTIs above 40% was dropped just before it was slated to go into effect, so that's to the benefit of all borrowers who carry high debt loads. For our part, we thought the introduction of a new credit tier into the LLPA matrix that sees borrowers with FICOs of 760-779 now subject to fees was more significant than were the other LLPA fee adjustments.

Meanwhile, the FHA did actually lower recurring annual premium costs for borrowers. trimming the annual percentage-based costs by 30 basis points across the board. No changes were made to the up-front mortgage insurance premium (UFMIP) or to MI cancellation policies. If the MMIF capitalization ratio continues to run at multiples of required levels, it's possible that some tweaks to those may come at some future point.

Underwater mortgages, again?

As noted in the FHA portion above, housing markets have cooled rapidly and appear to be likely to continue to do so for a while yet. Year-over-year home price increases are still very solid, but looking more closely at more near-term changes points to spreading declines in home prices that go beyond typical and simple seasonal patterns.

Data we analyze each quarter for our "Salary you need to buy a home in the top 50 metro areas" and Home Price Recovery Indexes use disparate data sets, one from the National Association or Realtors and one from the FHFA. Both have seen greater downturns in home prices in the third quarter of 2022 that are well above recent historic trends for the period. For example, the FHFA median home price data for the third quarter of 2022 showed 48 of the top 100 metro areas with lower home values than in the second quarter, By way of year-to-year comparison, in 2021 there was only 1 metro that saw this occur, just two in 2020 and only 13 back in 2019, pre-covid and pre-boom.

Using the NAR's data, we cover the top 50 metro areas, and 41 of those sported a lower sales price in 3Q22 than in 2Q; this was the largest number for this quarterly comparison by a wide margin looking back over the last seven years (the second largest was 28 metros in 2018). So while there's typically some seasonal softening for home prices in the third quarter of each year (and often more in the fourth quarter) it's also considerably more widespread this year.

That said, a quarter or even two isn't a exactly a pattern, let alone a trend, but given headwinds in the housing market, it may be the beginning of one. But is it troublesome?

For most homeowners who purchased at any time except the absolute price peak in a given market, any slump in home prices will simply reduce an equity position that expanded rapidly due to forces beyond their control; the market giveth and the market taketh away, so to speak. However, homeowners who bought their houses at or near peak for local valuations may find themselves in a position of no equity or even "negative equity", as home prices fall back from record highs to what in reality are only previous record highs.

Still, a decline in prices could expose some borrowers, especially those who only made small downpayments, such as FHA-backed borrowers (typically 3.5% down), HomeReady and Home Possible borrowers (3% down), VA- and USDA-backed borrowers (often 0% down) and even traditional conventional borrowers who put 5% down.

Even before prices began to soften, Corelogic estimated that 1.1 million mortgaged properties were underwater (1.9% of all properties with a mortgage) in the third quarter of 2022, a 0.4% increase from the 1 million and 1.8% rate in the second quarter. That small upturn isn't much, but could be an indicator of things to come in 2023. That said, the "negative equity" issue should remain contained for most areas and borrowers, but there could be concentrations of greater trouble in markets that saw outsized gains in prices over the last few years.

Mid-year review:
The softening in home prices is even more widespread than when we wrote this 2023 outlook. Looking at the FHFA's data for the first quarter (latest available), 22 of the top 100 housing markets now have median home values below the same period in 2022. Given that peak home values so far occurred in the second quarter of 2022, and that home price trends in 2023 have been broadly downward, it's a fair bet that there will be a lot more than 22 metros with annual declines when the second quarter data become available for review. Unfortunately, that's not until August.

As reckoned by median existing home prices across the top 50 housing markets (National Association of Realtors data) 23 of these metro areas saw lower home prices in the first quarter of 2023 compared to 1Q22. Like the FHFA data, it's very likely that a majority of the largest housing markets will post year-over-year declines in price for the second quarter. However, looking at the monthly trend for the median price of an existing home sold, existing home prices have been moving up from a January 2023 nadir, so the differential compared to last year has been small and comparisons will likely become more favorable once they no longer include last June's all-time peak for prices.

Although home prices increases have slowed, stopped or may have begun to reverse, the problem of folks owing more on their mortgages than their home is worth hasn't worsened significantly, at least so far. Corelogic reckons that in the first quarter of 2023, 1.2 million mortgage properties (about 2.1% of all homes with a mortgage) were underwater. While still a low figure, it is nonetheless trending in the wrong direction, if gently.

Home Equity / Cash-out refinancing

With homeowners sitting on deep levels of equity and home values still rising strongly, we expected to see a lot of cash-out refinancing in 2022, but interest rates rose so quickly that this option all but disappeared once 30-year fixed-rate mortgages climbed past about the 5% mark. That happened around mid-April, and mortgage rates would rise by another two percentage points by late October. As such, it made little sense for homeowners to replace a mortgage that was likely near all-time low levels with one with perhaps double their existing interest rate just to get access to a portion of available home equity. By October, mortgage analytics firm Black Knight noted that cash-out refis "accounted for just 4% of all prepayment activity ...] a record low." and through October 2022 that cash-out refis were down 83.6% compared to October 2021. Mortgage rates have since settled back appreciably since October's 20-odd year highs, but not nearly far enough to attract homeowners looking to tap equity via this financing method.

So cash-out refinancing activity is at a virtual standstill, as is rate and term refinancing. As long as interest rates remain above 5%, it will be hard to expect to see borrowers looking to tap equity willing to use a new first mortgage to get access to it. As far as 2023 goes, odds don't favor that we'll see mortgage rates get that low unless a severe recession gets underway, and even then probably not unless the Fed reconsiders its plans for reducing the size of its balance sheet.

All that said, and even in a higher interest rate climate, appetite for home equity loans and home equity lines of credit should continue to increase. Federal Reserve data covering commercial banks shows that outstanding home equity loans and lines of credit had been declining for more than a decade since a previous-boom peak, with loans retreating from as high as $1.138 trillion in the fourth quarter of 2007 (remember no-money-down piggybacks?) to as low as $389.6 billion by the first quarter of 2022. Now in the third quarter, volumes of home equity loans have picked up a little, rising to $419.2 billion, a 7.6% increase over the last two quarters, and with cash-out refinancing remaining low, there's probably space for another 10% increase or so in originations of home equity loans in 2023.

Home Equity Lines of Credit (HELOCs) followed much the same pattern, peaking back in the first quarter of 2009 at $736.2 billion outstanding, then falling for more than a decade to just $318.5 billion by the first quarter of this year. Now? They've climbed back up to $340.6 billion through the third quarter, an increase of almost 7% from the recent low water mark. Although the prime rate is likely to be at about 15-year highs in 2023 for at least a time, and HELOCs are priced a little above this index, homeowners looking to routinely tap portions of their home equity will likely to turn to HELOCS in increasing numbers. As with home equity loans, something on the order of a 10% increase in dollar volumes for next year seems most likely.

Mid-year review:
With loan rates still running at 15-year (or higher) levels, it's not surprising that refinance activity has come to a virtual standstill this year. Although we can contrive a few scenarios where some value can be realized from a cash-out refinance, the fact of the matter is that there is no value in doing so for the vast majority of homeowners, and won't be until mortgage rates are meaningfully lower.

High first-lien rates should be giving a significant lift to home equity borrowing, but that's not much been the case, either. Fixed-rate borrowing costs for second-lien products are above those for first-position loans, making them less enticing. The prime rate governs most HELOCs; it is currently at a 17-year high and threatening to be moved higher still, and the typical two percentage point margin added to it puts HELOC rates in the double digits for even good credit quality borrowers.

Still, and interest rate concerns aside, home equity borrowing is really the only game in town for homeowners. According to Federal Reserve data, closed-end home equity loans at banks increased to $435.7 billion through the first quarter of 2023, up by about 12% compared to the first quarter of 2022. However, activity is slowing, as the growth in loans has moved from a 4.07% increase in the second quarter of last year to just a 0.78% quarterly increase for the first quarter of 2023, so consumers may be becoming more cautious in their borrowing.

HELOC usage is also up reasonably, at least when comparing the first quarter of 2023 against the first quarter of 2022. HELOC balances have grown from $318.6 billion in Q1 last year to $348.9 billion, a 9.52% increase. However, like home equity loans, much of that increase came in the second and third quarters of 2022; after sporting increases of 3.66%, 2.96% and 2.26% in the last three quarters respectively, HELOC borrowing expanded by just 0.36% in the first quarter of 2023.

Our forecast for about a 10% increase in home equity borrowing seems on track for the moment, but of course there are three more quarters of data yet to be considered before 2023 finishes up.

Reverse Mortgages

Originations of HECMs outpaced our expectations in 2022, but it may be hard for them to do so again in 2023. A part of the reason that HUD reported a 31% year-over-year increase in HECM originations was due to HECM-to-HECM refinancing, as a lower interest rate creates a lower total future commitment that needs to be repaid when the homeowner exits the property, and refinancing using a new and higher property value can expand the amount of funds available for the homeowner to tap. However, as with forward mortgages, the current higher interest rate climate will slow this component of HECM originations, leaving only "traditional" and HECM for purchase purposes to drive the market.

That said, the traditional component does look pretty bright. Baby Boomers continue to retire, and by this measure there are likely about 4 million new retirees or so to be expected next year. While home values may soften up a bit in 2023, the 62+ year old homeowners looking for HECMs or reverse mortgages are likely to have been in their home for a longer period of time and so small wobbles in home values won't much change their ability to borrow. Moreover, it certainly doesn't greatly impact the reasons they would be taking a HECM in the first place.

In fact, the ability to tap what is a pretty deep pool of available equity as needed (or regularly) may prove very valuable to keep a retirement plan on track, especially at a time when stock and bond markets aren't exactly supporting retirement goals as they once had been. With no required payments, HECMs don't represent any commitment against income, so these funds can help tide over a rough patch for other investments for a fair period of time. In turn, this can help homeowners to be able to keep their holdings, rather than needing to sell them in an unfriendly environment, which risks both immediate loss and the loss of opportunity when equity prices eventually rise again.

In 2023, 'HECM for purchase' usage (which allows homeowners to buy another home using an HECM) will likely just perk along at low levels (due to still-high home prices in areas often favored by those looking to relocate and downsize in retirement). However, traditional income-support usage seems as though it will remain well supported next year, or at least until equity and bond valuations recover from their recent difficult climate. With the continued diminishment of the refinancing component, we think there will be a downtrend in overall HECM originations in 2023, which will probably settle back from the $64,437 billion tallied in HUDs 21-22 fiscal year to something closer to $55 billion, so perhaps a 15% decline.

Mid-year review:
So far for 2023, HECM borrowing is coming in considerably softer than we expected. HUD's fiscal year runs October through September; for FY 2022, some 64,457 HECMs were originated. Since October, just 19.960 HECMs have been originated; even if they run at the current average monthly rare over the next five months to complete the fiscal year, a total of 34,215 Home Equity Conversion Mortgages will be originated, so the decline from FY22 would be closer to 47% rather than the 15% we were expecting.

As is the case with first mortgages, refinancing of HECMs -- a process that allows a homeowner to increase borrowing capability by taking advantage of market-generated home price increases, and/or lower the accrual interest rate on the HECM -- has fallen to a near standstill. Last year in June, HECM refinances accounted for nearly 42% of HECM volume; As recently as April, that percentage has slumped to about 11%. Absent a sharp decline in HECM rates, which nearly doubled over the last year, HECM refinancing volumes will continue to be weak, depressing total volume.

Home prices

After significant increases over the past couple of years, home prices seem poised to settle in 2023. For some metro areas, this will mean smaller price gains, but for others, it seems as though some outright declines are more likely than not.

On a national scale (as though anyone could own such a property), values in the aggregate will like see a continued deceleration. Despite still-significant increases for much of 2022, the slowing in home price appreciation has actually been in place since the National Association of Realtors reported a 23.6% year-over-year increase way back in May 2021. By the time the calendar turned 2022, January's comparable annual increase was down to 15.3%, and by October, the median selling price of an existing single-family home was "only" 6.6% above the comparable year-ago level.

While there is no doubt still-considerable desire to buy a home, decreasing affordability due to high home prices and high mortgage rates are contributing to the significant slowing in sales. This put a damper on the kind of market conditions where folks were often paying above asking prices for available properties. That's really no longer the case, and with these headwinds already in place and the typical seasonal slowing in activity, there's a very good likelihood that home price increases will move to flatline in the next few months, at least on a national level.

We're already starting to see greater-than-seasonal declines in home prices -- and on a more widespread basis -- than is typical. Using National Association of Realtor data covering the top 50 metro areas, it's not uncommon to see up to half of the metros showing lower home prices in the third quarter than in the second of each year, and usually more metros show this pattern from the third to the fourth quarter. For 2022, 41 of the top 50 metros -- 82% -- posted lower prices in the third quarter compared to the second, and the national median home price for the period was also lower, too.

So more metros than usual saw lower home prices the later half of 2022. For markets that saw declines, the median value loss was 3.23%, the largest drop of any third quarter dating back to 2016 (second largest was 2018's 2.05% decline). As such, prices are softening on a more widespread basis than normal and in larger amounts than normal, and we haven't yet even seen all the data for 2022 as yet. It bears considering that the fourth quarter of the year often sees even more widespread quarter-to-quarter declines in home prices.

This sets the stage for 2023, though. The headwinds that have helped prices to start to soften -- high mortgage rates and high home prices creating affordability issues -- will likely self-correct to some degree during the year. Mortgage rates have recently come off about 22-year highs, falling back by more than a half percentage point, and this kind of decline presents opportunity for at least a few more potential homebuyers to participate. At the same time, slightly softer home values also serve to present the same opportunity, allowing a few more folks an opportunity to buy a home. These minor improvements in market conditions this can help stabilize demand, and in turn, also help stabilize home prices to some degree. Potential homebuyers wholly shut out of markets last year will rightfully treat these changes in conditions as opportunity and will jump in when they present themselves.

So this "opportunistic demand" should help keep home prices from falling too much, as will the other issue that has seen them skyrocket: a continuing lack of supply. Even with sharply diminished demand as 2022 rolled along, the supply of homes coming onto the market barely increased, and was actually lower in October 2022 (latest data) than it was both in September and a year prior. There remains little available to buy even with the decreased demand evident in the market, and this too will tend to help stabilize prices. That said, sellers that are actively in the market and most motivated (read: need to sell) may be willing to accept lower price offers.

We think that we'll see lower home quarter-to-quarter home values across all markets for the fourth quarter of 2022 and the first quarter of 2023. These declines may be sufficient as to create some negative year-over-year comparisons for some markets. After that, the spring season will re-lift prices as the normal surge of buyers comes into the market for the second quarter, but home prices will probably fade again in the third and fourth quarters, and some negative annual comparisons will show again. For the year, the most likely outcome seems like an overall slight decline in prices (or flat, at best) for most places and nationally, but some markets could see sharper resets for prices.

Overall, we don't expect to see the kind of value losses seen in the last housing downturn, but some folks that purchased in the mid-2021 to mid-2022 period may see the value of their home be less than they paid for it for a time. A recession would increase the odds of this, since struggling homeowners would be more likely to look to escape unaffordable housing situations, pushing more homes into the market./p>

Mid-year review:
At the moment, it looks as though our expectations for home prices have been pretty spot on. On a sequential, quarter-to-quarter basis, the 82% of markets with lower home prices in the third quarter of 2022 noted above gave way to 100% of metros with Q-to-Q reductions in the fourth quarter, but this figure dropped back to 78% for the first quarter of 2023 as prices started firming up again.

However, in the more widely-followed year-to-year comparison, the number of metros where the median price of a home sold is lower than a a year ago has been increasing. There was only one such metro with an annual price decline in the third quarter of 2022; this increased to 8 metros in the fourth quarter. For the first quarter of 2023, that number nearly tripled, with 23 metros sporting a median price lower than a comparable year ago period.

Declines aside, we also expected that prices would be well-supported this spring, and they have been. On a monthly basis, the median price of an existing home sold bottomed in January at $361,200; by May (latest month) it had risen to $396,100 -- a 9.7% lift off the January bottom.

Nationally, home prices did move to the flatline we expected; that happened in February. Since then, using a year-ago reference, this has turned into an outright downturn, and by May, the median national price of an existing home sold was 3.1% below the same month in 2022. As June 2022 was the monthly all-time peak for existing home prices, the coming June-June comparison (available late July) will likely show an even deeper decline. After that, year-ago comparisons may start to look more favorable, provided the demand for homes in the market remains fairly steady.

While home prices are firming now, we'll see if they re-fade again in the last half of the year, which is the second part of our expectation.