Erdmann Housing Tracker · Housing & Cities
TIER 5 Mon, 4 May 2026 14:03:28 +0000
I recently wrote about how the mortgage crackdown at the federal agencies in 2008 was, arguably, the overwhelming source of the sudden drop in spending in 2008 that was associated with collapsing nominal GDP growth, the deepening recession, and the financial crisis. ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- --- # Mortgages and the Supply of Safe Assets | | Kevin Erdmann --- | May 4 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- I recently wrote about how the mortgage crackdown at the federal agencies in 2008 was, arguably, the overwhelming source of the sudden drop in spending in 2008 that was associated with collapsing nominal GDP growth, the deepening recession, and the financial crisis. Figure 1 is my estimate of the quarterly net change in mortgages outstanding, by credit score, over that time. A stable economy in 2008 would have probably been associated with quarterly growth of around $100 billion or so in mortgages outstanding. | | ---|---|--- Figure 1 It occurs to me that there are associations here with the stock of safe assets, too. I wrote about safe assets in both of my books. But, I had failed to fully center the mortgage crackdown in 2008 as the cause of the downturn. I treated it more as a coincident or lagging factor. And, so I didn't fully account for its association with the supply of safe assets. Upgrade to paid In the post about the mortgage crackdown and aggregate demand, I noted that before 2006, when the trend in net mortgages outstanding in Figure 1 was still rising, those mortgages were associated with both funding new investments in housing and with creating liquid funds from home equity. But, "By 2006, mortgage growth was making up for _underproduction_. This played out regionally. And, by 2008, mortgage _contraction_ across the country was _causing_ underproduction." Where I wrote about mortgage growth and safe assets in the books, I discussed the demand for safe assets before 2008. Long-term interest rates didn't rise as the Fed pushed up its target short-term interest rate, and mortgage rates didn't rise either when they were trying to slow down construction. In the conventional literature, the focus is on chasing yield or on underestimating risks. But, those are just feeling different parts of the elephant that was a lack of safe assets. Long-term interest rates didn't rise when the Fed raised its target interest rate from 2004 to 2006 because savers were increasingly seeking _safety_. Before the Great Recession, there was rising demand for safe assets - central banks and savers around the world wanted to hold a lot of risk-free assets (like cash, Treasury bills, and other very safe debt). That's what was pushing down interest rates. And, by 2006, Americans were already reducing their ownership of real estate. Mortgage growth was still increasing, but the growth rate was declining fast. Unfortunately, the Fed interpreted that as if low rates meant credit markets were still stimulative to production, and that construction was collapsing anyway because it needed to. To the contrary, low rates and declining construction both pointed to risk aversion. Many savers wanted to fund low-risk borrowers and fewer people wanted to be owners and borrowers for them to lend to. This was the time of the CDO boom. This is when "The Big Short" was happening. So, the CDOs, and eventually the synthetic CDOs were a substitute for actual borrowers. Before that, mortgages were packaged up into securities, and arranged to make most of those securities very safe, and that met the demand for safe assets to invest in. Now, the investment banks had to start taking the less safe parts of those securities and recombining them so that some portion of them could transfer the risk through the new engineered financial products. More and more convoluted combinations were constructed in order to produce something that could be stamped "AAA". The conventional wisdom says that those mortgages in 2006-2007 were motivated by reckless borrowers and lenders, but it would be more accurate to say they were motivated by a lack of willing borrowers, and risk averse savers who found it increasingly difficult to _avoid_ risk as a result. That's the part I covered in the books. And, the kicker is that by 2008, all those mortgage securities weren't considered safe assets any more. There is a literature on the topic of safe assets and their part in the crisis story. Here is a table from a paper by Ricardo J. Caballero, Emmanuel Farhi, and Pierre-Olivier Gourinchas titled, "The Safe Assets Shortage Conundrum". The $20 trillion in safe assets in 2007 was already low enough to keep interest rates unusually low. And, when the mortgage pools lost their "safe" status, it dropped to $12 trillion. | | ---|---|--- The mortgage crackdown plays several roles here. First, I would argue that the mortgage crackdown was the _reason_ real estate collateral lost $5 trillion in value and led to the loss of "safe" status. Also, there was no reason for the GSE pools to lose "safe" status, especially after the government backed them. They were never insolvent. And, most of the losses that turned their net worth negative were prospective losses and accounting losses, many of which were later reversed, even in the terrible market they endured. But, conventional wisdom says the crisis was inevitable. Mortgage securities were destined to be unsafe the day they had been created - even the GSE securities. The crisis in 2008 and its aftermath had already been fated years before, and this was just a bubble running its inevitable course. That's the overwhelming and wrong tone of both academic and popular histories of the period. The same 3 authors also wrote another paper titled, "Safe Asset Scarcity and Aggregate Demand", where they describe how a lack of safe assets was actually slowing down growth after 2007. They write, "Acute safe asset scarcity forces the economy into a "safety trap" recession… Policies that increase the net supply of safe assets somewhere are output enhancing everywhere." But, as far as I can tell, they join the rest of the academy by simply treating this as a state of nature - something that couldn't or shouldn't possibly have been avoided. It simply is, and should be studied as it is. And, so there is the relationship between Fed tightening leading up to 2008, the mortgage crackdown's effect on spending in 2008, the mortgage crackdown's effect on real estate values and borrower collateral after 2008, etc. What if those mortgage securities could have still been safe in 2008 or 2011? The reason they weren't safe is that there was unanimity to benchmark public policy to their failure. The explicitly stated reason for monetary and regulatory policy choices in 2007 and 2008 was that if we didn't benchmark to deeply collapsing construction and home prices, we wouldn't have been properly disciplined. Even with the chance to learn from hindsight, that is the most common reaction I receive when I suggest that stable mortgage standards could have prevented low tier home prices from dropping to half their replacement value in cities across the country. The part I didn't fully consider in my books is that the mortgage crackdown made that problem even much worse. The world was crying out for safe assets, and by 2008, the federal mortgage agencies were telling $1 trillion worth of willing, qualified borrowers who could have been the catalysts for those safe assets, "No". When there aren't enough safe assets, interest rates get bid down. (Or, using my Upside-Down CAPM approach, the discount savers have to accept compared to at-risk returns increases.) So, the sudden annual loss of $1 trillion in borrowing, after borrowing had already voluntarily dropped by about $1 trillion, caused market interest rates to plummet. And, that is really bad news for a central bank that tries to stabilize the nominal economy by targeting interest rates. From 2009 to 2021, interest rates were stuck at zero. Table 1 shows that about $10 trillion was missing from the stock of safe assets, though GSE securities have returned to being treated as safe, and the Federal Reserve has used them as part of the large asset base it accumulated while it was trying to stimulate demand while interest rates were pinned at zero. | | ---|---|--- Figure 2 But, there is another $10 trillion missing. There had been a lending boom and a building boom that was associated with cyclically elevated home prices before 2006. By the end of 2007, the credit and cyclical impulses that I track with the Erdmann Housing Tracker model were both back to neutral. The boom had reversed. The net losses after that were due to the mortgage crackdown pushing home prices in low tier neighborhoods across the country well below a reasonable value. Figure 2 is a basic first-order approximation of what would have happened with mortgage levels and total real estate values in a world without the mortgage crackdown. The solid lines in Figure 2 are what happened. The dotted lines are what might have happened if we hadn't had a mortgage crackdown, but we still have other problems like the housing deprived Closed Access cities. The dotted lines assume that the discount from the mortgage crackdown and the rent inflation from the lack of home construction that followed it didn't happen. It assumes that home values grew with incomes after 2007 as residential investment led to the construction of the 15 million+ homes we are missing, and existing home prices increased with inflation. And, it assumes that mortgages outstanding would have scaled with home values, holding leverage stable over time. In that counterfactual scenario, there would be about $10 trillion more mortgages outstanding today. There would be $10 trillion more safe assets. The 2nd paper above by Caballero, Farhi, and Gourinchas was written in 2016. They concluded, "In the safety trap world, any policy that expands safe asset supply anywhere has expansionary effects everywhere and reduces risk premia. Public debt issuance is expansionary, to the extent that it is safe, and that future taxation does not curtail the private sector s ability to issue safe assets." I think they would say that the deficit spending during the Covid shock increased the real growth potential of the economy by releasing us from the "safety trap" of not having enough safe assets. Deficit spending created a lot of Treasuries for savers to invest in. In "Upside-Down CAPM" terms, low risk borrowers are the producers here, and the savers are the consumers. It is deferred consumption. They need someone to hold their capital for a while and then give it back to them. The US government is very good at providing that service. Possibly, also, the transitory inflation that was associated with Covid stimulus reduced the relative desire for saving versus investing, since it permanently put a 10% dent in the value of the consumption those savers were deferring. Figure 3 compares total safe assets (Treasuries + Mortgages) over time (Blue) or if Mortgages Outstanding had remained a stable % of GDP (Black). | | ---|---|--- Figure 3 What if there had been an extra $10 trillion of mortgage securities between 2007 & 2019\. What would the real yield have been on the 10 year Treasury? The lack of those mortgages has lowered real American incomes by reducing the real quantity of housing we have and increasing the scarcity rents families have to pay to stay in place. What if it also lowered our real incomes by keeping us in the "safety trap"? Maybe if there had been an extra $10 trillion in mortgages outstanding in 2015, the real yield on the 10 year Treasury would have been 2% and the Fed's large balance sheet and the Quantitative Easing activities wouldn't have been necessary because a Fed Funds rate above zero would have been stimulative in a market with more safe assets to invest in. Upgrade to paid You're currently a free subscriber to Erdmann Housing Tracker. For the full experience, upgrade your subscription. Upgrade to paid --- | | | Like --- | | Comment --- | | Restack --- (C) 2026 Kevin Erdmann 548 Market Street PMB 72296, San Francisco, CA 94104 Unsubscribe