Housing Affordability Depends on Building More, Not Cheaper Mortgages
Richmond Fed president Tom Barkin argues that housing affordability depends less on cheaper mortgages than on making homes easier and less costly to build, through changes to land use, permitting, redevelopment and local attitudes toward construction. In a discussion with Steven Davis at Jackson Hole, Barkin also describes a labor market with low layoffs but weak hiring, and says uncertainty over long-term real interest rates leaves fiscal policy with less reason to assume borrowing will remain cheap.
Housing affordability depends more on building than on cheaper credit
Housing affordability will not be solved primarily through cheaper mortgages or easier credit. Steven Davis and Tom Barkin instead put the emphasis on supply: land-use rules, permitting, construction costs, available land, and local willingness to allow development.
Barkin declines to prescribe government action on mortgage policy, but he points to several features of the current market that complicate the case for treating lower borrowing costs as the main remedy. Mortgage rates, he notes, are roughly where they were in 2004 and 2005, when housing was a vibrant market. Developers who say they are waiting for rates to decline will also acknowledge that projects do not pencil out for other reasons: construction and land costs have increased, permitting has become more difficult, uncertainty has risen, and projects require more equity than they did before the Great Recession.
The low-rate 2010s also changed the mortgage market. Barkin says adjustable-rate mortgages largely disappeared as rates fell, leaving fixed-rate loans at a share in the 90 percent range. Adjustable-rate lending was once a conventional way for households to make a purchase affordable. It remains, in his view, a reasonable part of a mortgage market, but borrowers no longer seem to want it to the same extent.
He also questions the assumption that homeownership is necessarily the best vehicle for building generational wealth. Across some five-, 10-, or 25-year periods he has seen, equity invested in the stock market produced better returns than equity invested in a home. Barkin is not advising households to abandon homeownership for equities. A house can be where people raise children, a family heirloom, and a personal touchstone. But it is illiquid, requires substantial upfront capital, and comes with turnover and hidden costs. Housing is not the only sector that might plausibly receive public support as a route to household wealth.
Davis argues that construction productivity has been poor for decades and that construction costs are exceptionally high in many parts of the United States. In his view, permitting and land-use regulation are major reasons. Making homes more affordable requires making them easier to build.
Barkin offers his Federal Reserve district as a practical comparison. North and South Carolina are among the states with the most housing construction, he says, while Maryland and the District of Columbia struggle more. The contrast can be visible across short distances: development appears just over the border from Maryland or Virginia in West Virginia, and south of Norfolk in North Carolina. The land is broadly similar, Barkin says; what changes is how it is priced, regulated, permitted, and encouraged.
You're competing to push them out, you're competing to bring them in, cause you want more housing.
The observation is aimed partly at small-town leaders who view developers with suspicion. Communities compete for development whether they acknowledge it or not. Restricting projects does not eliminate demand; it can direct construction toward jurisdictions where building is more feasible.
National homebuilders are still producing roughly 1.4 million residential units a year, Barkin says, but much of that building is occurring at scale on large plots of former farmland in West Virginia and exurban North Carolina rather than as dense infill. The resulting homes are not cheap in absolute terms. But homes priced around $300,000 to $350,000 are far less costly than comparable housing in high-cost markets such as California, and can put family housing within reach for buyers shut out of those markets.
Local governments have land, but must be willing to use it
Regulatory reform is not Barkin’s only supply-side lever. Tom Barkin adds imagination.
In communities that have lost population for years, blighted parcels and abandoned homes can remain outside the housing market because ownership is unclear or local systems are not designed to recover and reuse land. Land banks, he says, are one way to bring those sites back into productive use.
He sees other underused assets throughout his district. Churches are converting parking lots into senior-living facilities. Closed elementary schools can be repurposed for housing. Municipalities that say they want more affordable housing may themselves control land that could contribute to it.
Barkin once estimated that making housing affordable to a nurse required removing roughly 20 percent from the cost structure. Land, he says, is itself roughly 20 percent. The calculation changed what he noticed: potentially usable land was everywhere—in cities and towns, former school properties, parking areas, and neglected parcels.
The implication is not that every site should be developed, or that mortgage and credit conditions do not matter. It is that mortgage tweaks alone cannot sustainably overcome constraints in the availability, cost, and use of land. Places seeking more attainable housing have to lower barriers to construction and treat redevelopment as part of the solution.
A low-unemployment labor market can still be difficult to enter
Tom Barkin sees a labor market that is steadier than its headline unemployment rate alone might suggest, but not one that resembles the unusually tight conditions of 2018 and 2019. Unemployment remains low, layoffs are exceptionally limited, and earnings remain strong. Yet hiring is sluggish, wage pressure is muted, and new entrants—college seniors in particular, as Steven Davis notes—are having a harder time finding work.
The apparent tension reflects labor supply declining alongside labor demand. Barkin points to retirements and reduced immigration: the labor force is no longer expanding at the pace it did only a few years ago. Demand for workers has cooled as well, roughly in parallel. That helps explain why slower hiring has not produced a rise in unemployment.
The favorable side is unusually low separation from employment. New claims for unemployment insurance, measured against the size of the labor force, are at their lowest level since the series began in the 1960s, Barkin says. Other household-survey measures tell a similar story: flows from employment into unemployment because of job loss are very low.
The less favorable side is low turnover. Workers nervous about the outlook are less likely to quit for another job. Businesses worried that a long-predicted recession may eventually arrive are less likely to hire ahead of need. Those choices reinforce each other: fewer workers leave voluntarily, so employers need fewer replacements; employers hold back on hiring, so workers have fewer reasons to move.
That pattern can preserve employment without generating much dynamism. Firms do not have many excess workers to lay off, Barkin argues, while good earnings leave them with little immediate pressure to reduce headcount. But neither are they aggressively adding people. If earnings weakened materially, he suggests, the labor-market story could change. So far, it has not.
Inflation takes priority, but the Fed may leave more room for debate
Tom Barkin described Fed Chair Kevin Warsh’s Jackson Hole address as consistent with the principles Warsh has expressed to him. Barkin’s reading of the economic diagnosis was straightforward: labor-market conditions are in good shape, demand remains vibrant, artificial intelligence is a significant influence on that demand, and inflation remains too high.
Warsh’s inflation emphasis mattered most to Barkin. He cited Warsh’s reference to 65 consecutive months of inflation above target and his assessment that recent readings had been somewhat better than expected but still left substantial work to do.
Inflation's too high.
Warsh also rejected two ways of binding future policy: explicit forward guidance and a mechanical reaction function. Steven Davis took the latter as a signal that formulaic rules such as a Taylor rule may remain background guides rather than foreground commitments for policymakers.
Barkin’s interpretation focused more on the committee’s internal process. He believes Warsh wants the Federal Open Market Committee to enter meetings with room for a genuine debate and emerge aligned around an answer, rather than largely ratifying language negotiated in advance. Forward guidance can constrain that process by committing the institution to a future course before the relevant facts are known. A rigid reaction function could do much the same.
The current process, Barkin says, includes multiple rounds of drafting a policy statement before the meeting. Policymakers have opportunities to speak with the chair and work toward language that can receive committee support. The meeting itself is comparatively formal, with participants discussing economic conditions and policy through generally prepared remarks.
Barkin contrasts that approach with the Bank of England model described in Warsh’s 2014 review of the institution: a non-transcribed, free-flowing discussion of the economy before a subsequent day devoted to a policy decision. He does not say the Fed will adopt that model, but expects the Fed’s communications task force to debate the question. Mervyn King, a former Bank of England governor, is on that task force.
Barkin cautions against reading Warsh’s remarks as a commitment for the next meeting. Reporters repeatedly asked him what the speech meant for September; his response was that the chair had just renounced forward guidance. Markets and the media are still trying to understand what new leadership means for the Fed, he says, but the speech did not promise a particular policy outcome.
Uncertainty around neutral rates argues for fiscal prudence
Estimates of the neutral rate are too imprecise to serve as a practical policy lever, Tom Barkin says. The neutral rate, or r-star, is a key variable for monetary policy. But an estimate around 3 percent with a confidence interval roughly 150 to 200 basis points in either direction does not provide a sufficiently firm basis for setting policy.
That uncertainty framed Barkin’s response to Ken Rogoff’s presentation on rising long-term real interest rates. Barkin understood Rogoff to be arguing that the policy assumptions of the 2013–14 period—secular stagnation, persistently cheap money, expansive fiscal policy, and modern monetary theory—had been overtaken by an era of higher real rates.
The point was not that a particular future rate can be forecast with precision. It was that fiscal policy should account for the risk that rates revert toward longer-run norms. Long-term interest rates have been volatile historically, Barkin says, and policymakers do not fully understand when they will move higher. A chart of US interest rates over 250 years persuaded him that the unusual period was the 2010s, rather than current borrowing costs.
Steven Davis stresses the relevance of that uncertainty when debt relative to GDP is around 100 percent. If long-term real rates cannot safely be assumed to return to the low-rate 2010s, fiscal plans have less room to rely on cheap borrowing.
Warsh’s assessment of current financial conditions complicates the monetary-policy side of the question. Barkin notes that Warsh pointed to healthy financial conditions, the openness of the banking system, and investment in artificial intelligence as signs that the economy may not be meaningfully constrained. Barkin expects the committee to engage seriously with that proposition while inflation remains above target.