This is the second in a three-part series where I argue that mortgage rates are headed lower. It’s a more technical post—paying clients can text me directly with any questions here. The data from the charts and background reading is in the usual spot.

Twenty years ago I moved to India for the better part of a year, and kept a blog to record my adventures. I was re-reading the blog the other day and stumbled on this gem, from May 2004:

I tried to cut my hair. I did a great job on the front, but mangled the back. I had to call the bellhop—a young man I’ve befriended during my extended stay here at the cheapest hotel in Ahmedabad—to help me fix it.

He spent half an hour cutting away and accumulated a sizable pile of trimmings. When he finished, I turned around and saw his fearful expression. 

As he backed his way towards the door, I asked, ‘Is it good?’ 

His hand trembling, he pointed at me and replied, ‘There is blood’. 

The challenge of cutting your own hair is on my mind these days as I think about the “term premium”—the additional compensation investors require to tie up their money in investments for longer periods of time.

It’s an increase in this component of bond yields that had driven the selloff that began in the spring and ended two weeks ago. 

How much did the term premium increase? It’s hard to say: the term premium is like the back of your head: you can’t see it. But we know what happens when it rises—for bond investors, one might say “there is blood.”

If it’s hard to pin down the exact value of the term premium, we at least have a bit more clarity on what drives it. 

It’s driven partly by the balance between supply and demand for Treasury bonds. A soaring deficit is forcing the US Treasury to flood the market with bonds, and this may be one reason the term premium has risen. Some strategists have suggested international investors diversifying away from the US have dampened demand, exacerbating the effect of Fed selling as it unwinds its covid-era “quantitative easing” measures. 

The term premium is also proportional to uncertainty—the thicker the fog surrounding the future, the greater the compensation investors demand for lending into the unknown. The future is especially foggy right now: even as we enter the fourth quarter with the strongest growth momentum in 20 years (outside the pandemic), some of the most reliable leading indicators suggest a recession is imminent. Which one is it?

In a financial context, a “haircut” refers to a discount applied to the valuation of an asset to account for uncertainty. By the most common measure of the term premium—calculated daily by the New York Federal Reserve—the term premium is now over 0.4%. This is the highest it’s been since 2015, and represents a big haircut to bond prices for uncertainty. 

At this point, markets have digested the impact of higher Treasury supply, and even a small reduction in uncertainty will compress the term premium to more normal levels. It’s yet another reason to expect mortgage rates to drop below 6% over the coming months.

Mortgage interest rates track the yields on mortgage-backed securities (MBS)—which in turn track US Treasury bond yields, plus a spread. Mortgage rates are therefore largely determined by the same factors as US Treasury bonds: investor expectations about the path of Fed monetary policy, and a “term premium”.

This is the second article of a three-part series arguing that each of these components will decline, pulling down US Treasury bond yields—and mortgage rates. Today’s focus is on the term premium.

Here is the structure of the series and where we are now:

1. The Fed’s done hiking rates ←last week’s edition, available here

2. The term premium will fall ←today’s article

3. The spread between mortgage bonds and US Treasuries will compress ←next week

The term premium is the difference between the compensation that investors require for lending to the government for an extended period of time compared to what they would receive by rolling over short-term lending—a series of 1-year bonds—for the same amount of time.

Just as I offer a 17% discount for a one year subscription (here) compared to a monthly one, investors rightly demand some enticement to lock in lending for longer. The size of this enticement is the term premium.

Here is a chart that spells this out visually, though readers should beware the numbers are entirely made up (you’ll see why, below).

Defining the term premium is easier than measuring it. The problem is that, since it encapsulates the extra compensation for a lengthy loan over a series of short term ones, any estimate of the term premium hinges on knowing the price of these short term loans. But those lie in the future, and are unknown.

As researchers at the San Francisco Fed write: 

“Our uncertainty regarding the markets’ expectations of inflation and short-term interest rates in the far-distant future is reflected in our uncertainty regarding the current level of the term premium.”

Even if we can’t be sure of its exact level, all estimates of the term premium suggest it’s much higher than it was just a few years ago. The question for our purposes is—where is it going?

Here are two reasons I think it’s going lower.

Yes, the deficit matters

The government is running a gaping deficit, and it’s set to gape even more over the coming years. The aperture between revenues and spending has to be financed in bond markets. A deluge of bonds pushes up the term premium.

Three factors have recently pushed the deficit deeper into the red than earlier projected. 

First, tax receipts have been lower, partly because of stock market losses last year. The IRS also delayed the tax-filing deadline in several states affected by natural disasters, including California, until Oct. 16, after the end of the fiscal year. 

Second, spending on Social Security, which is indexed to inflation, has increased, as has spending on Medicare and Medicaid. 

And third, higher interest rates mean the Treasury has to spend more on debt interest costs. This is the main reason the deficit will grow over the next decade (excluding these costs, the deficit would basically remain negative, but stable).

Part of the problem is that, like a knocked-down boxer staggering to his feet only to be knocked back down again, we’ve suffered from a number of serious economic setbacks in a short period of time and have struggled to recover.

As the Congressional Budget Office (CBO)—the government’s budget watchdog—spells out:

In the 62 years from 1947 to 2008, primary deficits exceeded 2.5 percent of GDP only twice. In the past 14 years, they have exceeded that amount nine times—in large part because of legislation enacted in response to the 2007–2009 recession and to the recession of early 2020, induced by the coronavirus pandemic.

A price for everything

The Treasury will finance the deficit by issuing debt. The total stock of debt held by the public is forecast to rise from $26 trillion today to $46 trillion a decade from now. Many strategists are asking “who will buy it?” But, when it comes to US debt, a better question to ask might be “at what price?”

That’s because the US dollar remains the world’s reserve currency—the most important implication being that foreign central banks continue to store their funds in US dollars (and specifically, in interest-bearing US Treasury bonds). There is no reasonable worry that the US will default. 

But enticing buyers may require paying a higher interest rate (via the term premium). This is especially true because the biggest buyer of US Treasuries since the GFC—the Federal Reserve—has turned into a net seller.

The evidence suggests that current yields are plenty enticing. Foreign buying has been stable. And as the Fed has exited, households have entered.

If you’ve noticed the articles in the “personal finance” section of the newspaper pointing out the attractiveness of government bonds, you are not alone: households have steadily pulled money out of deposit accounts at banks to buy Treasuries as interest rates have risen.

This could continue. As my former colleagues at Barclays argue: 

…the household sector has been directly buying T-bills from the Treasury, and this process can go on for some time, as deposits are still elevated and are yielding well below T-bill yields.

If anything, the most recent news is positive: on Wednesday the Treasury announced it will be funding more of its debt in short term bills rather than 10-year notes—this relieves some of the pressure on that part of the interest rate curve (the part of the curve that mortgages track most closely). Over the past two days, mortgage rates have plummeted at the fastest pace since March.

The bottom line is that concern about the amount of debt the Treasury needs to finance isn’t overblown, merely that—after the carnage in the bond market over the past few months—that concern is now reflected in the price. 

Uncertainty is priced in, and is likely to fall

I’d make a similar argument about the effect of uncertainty on the term premium. There are good reasons for it, but it’s already in the price. 

This was not the case a few months ago—back then, uncertainty was high, and the term premium hadn’t yet adjusted.

But, again, the calamitous performance of bonds over the past few months have cheapened them to a point that investors are now being fairly rewarded for uncertainty.

Barclays, again:

“with the sell-off in the long end, we see the term premium as having been restored to reasonable levels.”

Of course, uncertainty could remain high, or even climb from here. This would, for example, happen if inflation reaccelerated. But I think the opposite is much more likely.

Uncertainty about interest rates is most pronounced at turning points in Fed policy cycles—and that’s exactly where we are now. The Fed held rates steady this week. If they hike in December (markets are pricing a 50/50 chance), it will almost certainly be the last hike of this cycle. As Fed policy rates crest and stabilize, uncertainty should decline—and with it, the term premium.

Conclusions

One reason mortgage rates are so high is because a component of US Treasury yields—the term premium—is unusually elevated.

As markets get comfortable with Treasury issuance, and as uncertainty abates, the term premium—and mortgage rates—should decline.

Along with the change in Fed policy I wrote about in part one of this series, a declining term premium is another reason I think we are headed towards sub-6% mortgages a year from now.