Exorbitant Exorbitant Privilege
June 20, 2023
Temporary economic stimulus from impact of higher rates on mark-to-market homeowner balance sheets
Few things in economics are as widely known as the stimulatory effects of lower interest rates and the inhibitory effects of higher interest rates. And so with this first piece, I would naturally like to challenge this assumption as false, albeit only in the particular circumstances of a rare corner case – but as it happens, a corner case which we are currently experiencing.
In ordinary circumstances, higher interest rates increase the cost of borrowing, making it harder for businesses and consumers to spend or invest. Companies down-size their financial plans, households reduce their spending, and each reduction perpetuates the slowing of economic activity in a self-reinforcing mechanism. Higher rates slow the economy by design. Standard econ 101.
However, I’d like to argue that in the specific corner case of: (1) record low interest rates (2) followed immediately by a record steep rise in interest rates in (3) a housing market financed with predominantly long-duration fixed-rate mortgages, the effect of the rise in interest rates is temporarily stimulative of the economy. Homeowners actually are (and feel) wealthier during the rapid rise in rates, boosting consumer confidence and stimulating the economy in a way that counteracts the otherwise inhibitory effects of higher rates. Up is Down. I believe this corner case is applicable to the United States and helps explain the surprising strength of the US consumer during the period of rising rates in the past 4 quarters.
Homeowner balance sheet
The mechanism of action for this temporarily stimulative effect of rising rates is positive impact on homeowner balance sheets.
Approximately 65% of American households own their primary residence, and this residence is typically their largest asset, dwarfing the rest of their balance sheet. Financing this asset in the US is typically a 30-year fixed-rate mortgage at 80% loan-to-value. As of this writing the median home price in the US was $463,000. The homeowner’s balance sheet would look something like this:
The 30-year fixed-rate mortgage is a great privilege of the American consumer, thanks to the status of USD as the world’s reserve currency. I am aware of no other consumer base in the world (whether Canada or Europe or anywhere else) with access to anything remotely similar. Moreover, during 2020/2021, US consumers were able to refinance their mortgages with record-low rates of around 2.5% to 3.0%, fixed for 30 years. Wisely, many American homeowners did exactly that. According to New York Fed data, there were approximately $8.3 Trillion of purchases or refinancings during the 2 years since the start of Covid-19 in the US, compared to $3.9 Trillion in the 2 year period before.
Bond math
A 30-year fixed-rate mortgage is a bond (like a 30-year Treasury Bond), and like any fixed-income instrument has a face value and implied market value that behaves in accordance with bond math. Consider a 30-year bond with a face value of $100 and an annual coupon of 2.5%. At issuance when market rates are 2.5%, the bond value is of course $100. When rates have risen to 6.7% (the current mortgage rate), the bond value drops dramatically to a mere $46. This is hard to overstate. The 390 bps increase in rate results in a 54% drop in the value of a 30-year bond.
For the median home of $436,000 with a 80% LTV mortgage, the $348,800 principal balance of the mortgage financed at a 2.5% fixed rate for 30 years drops in real value by 54% when rates are at 6.7%. The banks and lenders that loaned that money are now holding on to a bond only worth $160,448. For the fortunate American homeowner, although on paper they still owe the face value principal balance, the economic amount owed is the much more attractive $160,448.
What about home prices, the asset side of the homeowner balance sheet? The prices of homes has not fallen. According to the St. Louis Fed, the median home price in Q1 2022 was $433,100, effectively the same (indeed a touch lower) than what they are today.
Mark to market
The homeowner balance sheet when rates went from 2.5% to 6.7% therefore results in a dramatic increase in mark-to-market homeowner equity:
The median US homeowner has experienced a dramatic $188K increase in net worth while mortgage rates went from 2.5% -> 6.7%, through erosion of the value of their mortgage liability. In aggregate, we mentioned above that $8.3 Trillion of purchases & refinances occurred during the 2-year Covid-19 period. The same bond math implying a 54% drop in valuation would imply a $4.5 Trillion positive impact for American households.
Real impact
At first glance, one might dismiss this as merely theoretical, an accounting nuance. Do homeowners understand bond math, and would they behave differently because of mark-to-market calculations they have never seen? Directionally at least, homeowners have absolutely behaved as if they received such a stimulus as rates rose.
Homeowners feel fortunate to have their ultra-low mortgage rates, and they think about it as they consider their finances, feel the impact of inflation on food & other purchases which contrasts so starkly with their wonderfully fixed mortgage payment. They’re reluctant to sell their house even when life circumstances change, eager to hold on to their mortgage rate. And they’re behaving with a resilience and confidence in both the labor market and as consumers, in a way that has surprised the markets and the Fed during the past year of rate increases. It is less surprising when we realize the positive mark-to-market impact to homeowner balance sheets as their largest liability shrinks dramatically in value.
Caveats
This stimulus of $188K per homeowner is a gross impact, not a net impact, from the increase in rates. Felt inflation at grocery stories and throughout the consumer economy has a dampening effect which erodes the stimulative impact of the stimulus of mark-to-market mortgage value.
Mark-to-market impact is a zero-sum game for the economy, so while homeowners reaped the benefit of the fall in the value of their mortgage liabilities, there was an equal and opposite loss from the lenders. We saw the beginning of this a few months ago in the failure of Silicon Valley Bank and particularly First Republic (whose asset base was ill-advisedly comprised almost entirely of fixed-rate mortgages). Indeed, the failure of banks and the market to comprehend the difference between face-value and mark-to-market value of these fixed-income instruments was the crux of the failures. The stimulatory effects for homeowners is simply the other side of the same issue.
When rates stabilize, at whatever level, the stimulus goes away, leaving only the inhibitory impact of the higher level of rates. This is because the stimulatory effect on homeowner balance sheets via bond math is not from higher rates (i.e. the level of rates) but from the rapid increase in rates (i.e. the pace of increase) relative to the low-rates when the mortgages were originated.
The effect described is due to the 30-year term of mortgages in the US, which exaggerated the impact of the record-low 2.5% mortgages rates when they were offered. Almost all refinancing demand for the foreseeable future was pulled-forward during this time, and locked in for 30 years. As such, any reduction in interest rates (i.e. Fed attempts to stimulate the economy if we fall into slowdown in the coming years) is likely to be relatively muted as the mechanism of action from the housing sector has already been exhausted.
This is a US phenomenon where 30-year fixed rate mortgages are possible & common. Outside of the US, such an attractive product is simply not possible – precisely because of the risk to lenders described here.
Near-term implications
If the analysis of this impact is correct, I believe it would imply the following near-term differentiated conclusions:
The unexpected strength of the US consumer and labor market in the face of the rapid rise of interest rates in the past 12 months was temporarily boosted by the mark-to-market reduction in homeowner mortgage liabilities, which will disappear if/when mortgage rates stabilize or fall
Near-term future recessions will be riskier as they will be harder to counteract with monetary policy since record-low rates resulted in pull-forward of potential refinancings at rates fixed for 30 years
Financial system instability is of heightened risk beyond the relatively muted disruptions from SVB, Signature, and First Republic. The mark-to-market reduction in mortgage values was a transfer of wealth to homeowners, and the effect of this lost value on the lenders is not complete
Interest rates may need to remain at an elevated rate (but not necessarily rising) for many years to allow the economy to digest the dramatic pull-forward in the housing market. Economists and monetary policy experts may do well to pay attention when extreme, untested policies create unintended distortions and wealth transfers that can have significant effects on the real economy and across society