Theory of the Term Structure

The term structure of interest rates or the yield curve maps bond maturities to rate bond yields. Imagine a world where only two maturities of bonds exist, one and two years. Investors must allocate their portfolios between these two maturities. They solve this allocation problem by recognizing that the broad asset market equilibrium ought to bind. That is, it is impossible to earn systematically higher risk adjusted returns by holding one maturity of bonds over the other.

In general, this condition will hold when the two year yield is equal to the average expected value of the one year yield over the next two years. We can think of this average as the one year yield today plus the one year yield one year in the future, divided by two.1 One of these terms is directly observable, the one year yield today, the second is not, the one year yield one year forward. However, we can use the two year yield to infer market expectations about the forward yield on the one bond.

If the two year yield is significantly higher than the one year yield, the market expects interest rates to rise, since it signals a higher expected one year forward rate. If the two year yield is below the one year, that is the yield curve is inverted, it signals an expected decline in short term interest rates. Market prices for bonds longer term bond yields are thus intimately linked expectations about the future path of short-term interest rates.

There is No “Free Market” Determining Bond Yields

Short term interest rates in the United States are set by the Federal Reserve, the nation’s central bank. But since bond yields are an expression of market expectations about future short-term rates, bond yields are in the end just a market signal about the Fed is expected to do.

Put another way, Treasury market pricing action does not reflect anything about “fundamental” or “correct” levels for interest rates. Rather, it reflects market participants’ expectations about future interventions.

So long as we have a fiat money system, the Fed cannot therefore “step back” from the market. Under its current framework, it must set interest rates somewhere, and wherever it sets them, it will have a huge impact on bond yields.

In this light, rolling back forward guidance to let the market “sort out” the “correct” level for bond yields is terribly misguided. As Grep Ip of the Wall Street Journal put it, the Fed is not the umpire there to call balls and strikes. Its the most important player on the field. The practical effect of reduced forward guidance is to simply increase term premiums, of the compensation investors demand to hold longer dated securities. Increased uncertainty about the path of short term rates increases the market risk of holding bonds, since bond prices adjust to changes in the expected path of short rates. All this does is drive up borrowing costs across the economy. In other words, cancelling forward guidance is back door monetary tightening in a imprecise and unwieldy manner. Raising borrowing costs are the Fed’s main tool to fight inflation, but when it uses it, it should be done in a transparent and straightforward manner.

Conclusion

The Federal Reserve cannot plausibly separate itself from the determination of long-term interest rates. Long-term Treasury yields reflect the expected path of short-term rates, together with compensation for the risks associated with holding longer-duration securities. Because the Fed determines the current short-term policy rate and heavily influences expectations about its future path, its policy framework necessarily plays a central role in determining the entire yield curve. There is no independent “market” rate of interest that emerges after the Fed simply gets out of the way.

This is why greater ambiguity about future monetary policy should not be confused with greater reliance on markets. Markets must still form expectations about what the Fed will do; they simply have less information with which to do so. The resulting increase in uncertainty is likely to appear as a higher term premium, greater bond-market volatility, and consequently higher borrowing costs for households and businesses. Forward guidance can certainly be overused, and the Fed should not pretend to know its future policy path with certainty. But communicating its reaction function and reducing unnecessary uncertainty are themselves important tools of monetary policy. A central bank that deliberately obscures its intentions does not relinquish control to the market. It merely makes the market’s attempt to anticipate the central bank more costly.

Commuter Decisions to Use Roads

The decision to use public roads can be analyzed in a simple cost-benefit analysis framework, just like any other economic decision. Commuters will only use a road if the benefits outweigh the costs. The benefits derived from travel are whatever the utility one gets from accessing their destination. Costs include travel time, risks to personal safety, fuel, wear and tear on one’s vehicle, insurance, and other costs. Every time a commuter receives more benefit from travel than the cost, welfare is enhanced.

When a new road is built, it can increase welfare in different ways depending on how commuters respond. If the increased road space does not induce more driving, existing commuters benefit via reduction in travel time. If commuters respond by driving more, and congestion is not reduced, aggregate welfare is still enhanced, because more commuters get to consume road space at a similar cost.

The Futility of Building?

Some urbanists imply that building new highways is a waste of time because studies like the one above show that new roads usually don’t reduce congestion. However, this ignores the benefits that new drivers derive from expanded roadways. Ironically, this argument has strong resemblance to NIMBY assertions that new housing construction is futile, since new housing may simply attract more in-migration and thus not lower housing prices very much.

The key point is that even if we accept these outcomes as given (they are not), it is an argument in favor of more building (roads and housing) not less. After all, what would be the point of building new housing that just sits vacant or roads which are not utilized?

The Saliency of Certain Benefits

Fundamentally, these analytical errors come from the fact that certain benefits are more politically salient than others. On the housing front, people want lower prices, especially because of the big run up in housing prices since 2019. Thus, arguments that increased carrying capacity in cities in beneficial even if it does not reduce housing prices very much (or at all) fall flat. Similarly, drivers are a large political constituency and they ultimately want shorter commutes with less traffic. When expanding roadways only results in more people being able to travel without reducing travel times, it enhances aggregate welfare, but it does not help incumbent drivers. Thus, these kind of aggregate welfare arguments won’t gain much political currency.

Conclusion

The lesson from induced demand is not that building roads is futile, but that congestion reduction is an incomplete measure of their value. A roadway that accommodates substantially more trips without producing faster travel may still generate significant benefits by allowing more people to reach jobs, businesses, homes, and other destinations.

The broader mistake is to judge increases in utilization only by whether they improve conditions for incumbent consumers. New housing need not dramatically lower rents to be valuable if it allows more people to live in a desirable city. Likewise, new road capacity need not permanently reduce congestion to be valuable if it allows more people and goods to travel. In both cases, greater utilization is not evidence that additional capacity has failed; it is often evidence that the capacity was valuable enough for people to use it. Public policy should therefore focus on whether the benefits of additional capacity exceed its costs, rather than treating the failure to achieve one especially salient outcome—lower rents or shorter commutes—as proof that building was pointless.

Banks are fundamentally in the cash flow business. They purchase future cash flows from customers who sell them in exchange for deposits which can be spend immediately. This phenomenon, known as maturity transformation, is not limited to banks.

A humongous market for cash flows exists in the form of all kinds of securities. A bond is just a stream of coupon and principle payments and large financial institutions invest in bonds using leverage. Much like the institutional framework between banks and the Fed, large players in the US bond market always have access financing because of Fed policy. The constraint on their ability to invest in bonds is thus not constrained by some finite pool of liquidity, but rather market imposed capital constraints such as haircuts on repo transactions.

These firms face a simple choice when entering the market. If they meet market imposed net worth requirements and long bonds offer enough spread over short rates to compensate for duration risk, they should borrow short and invest. Much like banks, the limiting factor is capital and the availability of investable securities, not liquidity.

Thus, when thinking about asset price formation, we must not fall prey to “quantity” theories of supply which predict that increased bond issuances must push up yields. A well capitalized finance sector can easily absorb reasonable increases in the supply of Treasurys without yields rising (that is, bond prices falling).

This brings me to what I call the marginable capital theory of asset prices. As long as financial capital exists which can be offered as margin to get loans to invest in assets with predictable cash flows, an increase in the supply in those assets won’t affect prices. It will only draw in existing leveraged investors into the marketplace. A well capitalized investment banking sector will always do its job and engage in maturity transformation in the securities market!

The financial sector could suffer a large net worth shock, such as in 2008. This could limit the the ability of bond dealers to enter the market as new assets are emitted, and thus long term interest rates could rise relative to short rates, as well as credit spreads and other risk premia. If the normal activity of the financial sector can no longer absorb the incremental supply of new assets, yields would have to rise so that the asset market clears.

Alternatively, a massive surge in new issuance, maybe on the order of 10T or greater in a very short period, could force yields up because investment banks simply wouldn’t have the capital to absorb the new supply at current prices. This scenario is academic however, the Fed could easily intervene, or the Treasury could avoid the situation all together by issuing at the short end of the curve.

In any event, supply is really only an issue in extreme and/or dysfunctional markets. Assets are cash flows and its usually profitable to invest in with leverage. In the case of bonds, prices are set by expectations about the present value of the underlying cash flow, or in other words, expectations about interest rates. Mechanically, well capitalized firms are easily able to access the leverage they need to absorb new issuances without pushing prices down. The issue isn’t supply or loanable funds. It’s marginable capital

Nowhere in the country is this better illustrated than in New York City. Through a convoluted system that favors homeowners over apartment buildings, New York City imposes some of the highest effective tax rates on large multifamily rental buildings in the country. While New York City is currently experiencing a building renaissance thanks in part to a recently enacted tax-abatement program, the city must address the underlying dysfunction in its property tax system to truly unleash its building potential.

The Impact of Property Taxes on Development Economics

The viability of development ultimately rests on the extent to which the value of a completed project exceeds its cost. Apartment buildings are valued based on the income stream they produce. A typical newly constructed apartment building is worth about twenty times its annual net income. This can also be expressed through the building’s capitalization rate, or cap rate, which is simply net operating income divided by value. For a building trading at twenty times its net income, the cap rate is 5 percent.

For a building with a 5 percent cap rate, every $1 increase in NOI increases the building’s value by $20. This is because $1 × 20 = $20, or $1 ÷ 0.05 = $20. Both calculations are equivalent. Already, we can see how even small changes in net income produce large changes in building value. This is why development is so sensitive to rent levels and rent growth, as well as to large increases in expenses such as insurance and property taxes.

To drive this point home, let’s look at a toy example of a large apartment complex that could be constructed for $80 million and sold for $100 million upon completion. Suppose the completed project has a 5 percent cap rate and will face a property tax equal to 1 percent of the building’s value, or $1 million.

Here, the proposed project has a decent margin for the developer. But let’s see what happens if the property tax is increased to 2 percent.1 This reduces the net income the building produces by $1 million because the property tax has increased from $1 million to $2 million. The new net income is therefore $4 million. If we divide that amount by the cap rate of 0.05—or, equivalently, multiply it by 20—we calculate a building value of $80 million, equal to the cost of construction. The project’s entire profit margin is wiped out.

New York’s Dysfunctional System

New York City’s system actually begins with a reasonable premise. First, the city determines how much revenue it wishes to raise through property taxes in a given year. However, the process quickly goes downhill. The city divides its properties into four classes:

Class 1: Homes with three units or fewer and some condominiums

Class 2: Most apartment buildings, co-ops, and condominiums

Class 3: Utility-owned property

Class 4: Commercial property

The city then uses an unusual system to assess taxes. In most jurisdictions, each property receives an assessment and then pays a tax rate multiplied by that assessment, usually between 1 and 2 percent. In New York City, each class of property has both an assessment ratio and a tax rate. The assessment ratio determines the share of the property’s calculated market value that is subject to tax. The tax rate is then applied to that assessed value.

Class 1 properties have an assessment ratio of just 6 percent. The tax rate is 19.843 percent. Multiplying the two produces an implied tax rate of approximately 1.191 percent of the property’s calculated market value. Thus, the implied tax rate on small properties such as single-family homes and duplexes is relatively low. The effective tax rate faced by Class 1 property owners can be even lower once exemptions, abatements, and assessment caps are considered.

Multifamily properties have a considerably higher assessment ratio of 45 percent and a slightly lower tax rate of 12.439 percent. The math produces an implied tax rate of approximately 5.6 percent of calculated market value, a considerably less favorable result than for single-family homes. After abatements, the effective tax rate can be somewhat reduced, but unabated buildings may face effective tax rates of around 3 percent of market value or more.2 As we saw in the previous section, rates this high can crush development economics by substantially lowering the value of a building once it is completed.

A Brief History of New York City Tax Abatements

Because new developments face incredibly high property tax bills, New York City has a long history of using tax abatements to spur construction. The first major program, 421-a, was passed in the 1970s, and its original goal was to maximize construction rather than produce affordable housing.

At the time, New York City was in crisis. It was teetering on the edge of bankruptcy and experiencing significant population loss. Any investment was welcome, and the city sought to encourage it through tax incentives. The original program was quite generous. The value added to a parcel through construction was exempt from taxation, while the owner continued to pay taxes based on the original value of the land.

As New York City recovered, many came to believe that the program was too generous and was simply subsidizing projects that would have been built anyway. Over time, the city made affordability requirements a condition of qualifying for the program. The program was ultimately allowed to expire in 2022, which caused some proposed projects to be abandoned.

In 2024, a new tax-abatement program was created. Officially called Affordable Neighborhoods for New Yorkers, it is known in the real estate industry as 485-x. The program restores tax abatements for new projects in exchange for affordability requirements. Depending on the size and location of the development, approximately 20–25 percent of the units must generally be affordable. The program also imposes labor standards on projects with 100 or more units.

New York City is now experiencing an increased level of construction, a notable achievement because it bucks the national trend. Nevertheless, the city will not reach its full development potential until it reduces the heavy tax burden imposed on apartment buildings. Although the latest tax-abatement program helps some projects pencil, developers must incur other costs to qualify for it. As the preceding math demonstrates, increasing a project’s potential NOI can dramatically increase the value of the completed building, providing rocket fuel for development.

Conclusion

New York City’s property tax system imposes an unusually heavy burden on apartment buildings, directly reducing their net operating income and, through capitalization, reducing their value by many times the annual tax bill. For new development, this can mean the difference between a project generating an acceptable return and not being built at all. The city’s high taxes on multifamily housing therefore do not merely reduce the profits of existing landlords; they also discourage the production of the new housing New York desperately needs.

For decades, the city and state have attempted to compensate for this problem through tax abatements such as 421-a and, more recently, 485-x. These programs can make development feasible, and the recent increase in construction demonstrates how strongly builders respond when the tax burden is reduced. But abatements are an unnecessarily complicated solution to a problem created by the underlying tax system. They require developers to navigate detailed affordability, wage, and eligibility rules simply to obtain relief from taxes that would otherwise make many projects uneconomic.

A more durable solution would be to reform the property tax system itself. New York should reduce the disparity between the treatment of small owner-occupied properties and large rental buildings, lower the penalty imposed on new improvements, and create a more predictable tax burden for multifamily housing. Any reform would require difficult choices about how the city raises revenue, but the current system is neither equitable nor economically efficient. New York cannot simultaneously claim to face a housing shortage and maintain a tax structure that sharply reduces the value of constructing apartments. Tax abatements may keep some projects alive, but only structural reform can fully unlock the city’s housing-production potential.

Basic Rent Control Economics

Rent control misallocates housing via lock-in effects. Not only do landlords lose out on higher profits, but occupants cannot maximize the utility of said housing. An apartment next door to a hospital may be highly valued by a doctor but much less valued by random person who works on the other side of town. Allowing the doctor to outbid other prospective tenants not only raises the landlord’s profits, but it actually puts the apartment to its “best use” as expressed via the doctor’s high willingness to pay. Similar arguments can be made with respect to unit size. For example, a family of 5 would probably be willing to pay significantly more for a three-bedroom apartment than a single person would. By allowing open bidding on housing to occur, the market naturally sorts housing into its “best use” via the price system.

Rent control ends this market-based sorting and reduces both landlord profits an aggregate consumer welfare. However, it comes at the benefit of stabilizing housing costs for tenants in rent stabilized units. Economics as a discipline cannot determine whether this tradeoff is worth it. It is largely a political question. Economics can only measure the aggregate welfare lost from rent control and the level of de facto redistribution that rent control enacts.

Fixed-Rate Mortgages

The dominance of fixed-rate mortgages is unique to the United States and only made possible via government subsidies and guarantees. Just like rent control, it leads to a misallocation of housing. Homebuyers who are lucky enough to buy when rates are low get to lock-in their borrowing costs for up to 30 years. When interest rates rise, moving means getting a new loan at a higher rate, which can significantly increase housing costs. However, the circumstances of households frequently change such that a move may be optimal but for higher borrower costs. For example, a change in job, or an increase in family size may motivate a move. But since many households have locked-in mortgage rates well below current mortgage rates, many choose to just stay put. This distorts the housing market at the expense of new entrants, who have fewer potential options to choose from when they go looking to buy. It also reduces mobility by widening bid-ask spreads. An existing homeowner with a locked-in low rate mortgage will have a much higher reserve price to sell their home. But since a new buyer faces higher borrowing costs, their ability to pay these high prices is greatly reduced. This depresses home sales and causes households to stay put longer.

This reduced mobility also distorts the labor market. There is evidence that households with low, fixed-rate mortgages decline to move even when presented with better employment opportunities because they do not want to lose their low rate mortgage loan. Similar employment effects have been documented with rent control. Since moving means losing a coveted low rent apartment, tenants in stabilized units significantly narrow their job searches and are thus more likely to remain unemployed for longer.

The Politics of Housing Stability

Both rent control and fixed-rate mortgages distort the housing market. About half of the 50 million fixed-rate mortgages in the United States are below 4 percent, a steep discount compared to the going rate of about 6.5 percent. At the same time there are about 2.5 million rent stabilized units in the entire country. With roughly ten times more households with “locked-in” mortgages than there are households in rent stabilized units, there can be no doubt about which policy is more distortionary. However, relatively little outrage is directed at US mortgage policy which makes fixed-rate loans possible in the first place. The reason for this is politics. Homeowners are a powerful political constituency and households highly value the ability to lock-in their housing costs.

As a society through our elected representatives, we have decided that greatly warping the for-sale housing market (and the labor market) is a price worth paying to stabilize costs for incumbent homeowners.1 Relatedly, to further appreciate the exceptionally high value that households place on stable housing costs, one can look no further than the property tax revolt playing out across the country.

It should come as no surprise therefore that tenant groups are organizing tenant political power to obtain the same stability that we give to homeowners. Yes, even at the expense of prospective tenants and large distortions to the rental market.

Conclusion

The point is not that rent control is costless, or that economists are wrong to identify its distortions. Rent control really does reduce mobility, misallocate units, reduce the number of available units for rent at any given time, distort the labor market, and redistribute welfare toward incumbent tenants. But those critiques apply with surprising force to another form of housing-cost stabilization that America not only tolerates, but actively subsidizes: the long-term fixed-rate mortgage.

A homeowner with a 3 percent mortgage is, in practice, protected from the market in much the same way as a tenant in a rent-stabilized apartment. Both receive a valuable incumbent benefit. Both face a large penalty for moving. Both may remain in housing that no longer fits their needs because leaving would mean surrendering that benefit. And both systems impose costs on outsiders: prospective tenants in one case, prospective buyers in the other.

The difference is not economic purity. It is political power. Homeowners are treated as prudent families deserving stability; renters are treated as greedy market distorters when they demand the same thing. But if housing stability is valuable enough to justify structuring the entire American mortgage system around it, then it is hard to argue that renters are irrational or uniquely destructive for wanting similar protection.

The honest debate is not whether housing-cost stabilization creates distortions. It does. The honest debate is whose stability we are willing to protect, and why the largest incumbent subsidy in American housing policy is rarely described as a subsidy at all.

MMT has long been a punching bag for mainstream economists, and heterodox economists have punched back very hard too. Despite the heated discussions, the great irony is that the ideas from MMT fit perfectly fine within mainstream macro models. MMT, at its core, is actually just standard macro when certain conditions are met. Specifically, MMT models fully remove any nominal solvency constraint on governments which print their own free floating currency. Under this condition, the need for an interest rate instrument goes away, since a single policy variable (the budget deficit or surplus) is sufficient to stabilize the level of inflation/output gap.

Functional Finance

MMT as defined by actual heterodox economists, not online posters, is very easy to understand. Its intellectual antecedent is functional finance, an idea developed by the American economist Abba Lerner. Lerner observed that governments which print their own currencies and allow them to float freely faced no nominal budget constraint. Or in other words, in theory the government can always just print money to extinguish any nominal liabilities it has. This already scares many people and shuts down the conversation. But this observation should not be controversial. It is not a prescriptive policy recommendation.

The observation is important because of what mainstream macro has to say about the government’s budget constraint, regardless of monetary regime. So-called “sound finance” says that the present value of future primary surpluses (the budget balance excluding interest costs) must exceed the face value of the current debt stock.

In a functional finance world, the nominal budget constraint does not exist. Only a real budget constraint does. The government cannot buy what is not for sale no matter how much money it prints. Once it increases the deficits past a point where the economy has reached its productive capacity, it merely bids up prices. When no solvency binds, the budget deficit becomes solely a tool to target the level of inflation and close the output gap (when one exists).

Standard Macro

The key way standard macro differs from MMT is that it has a large role for monetary policy which is distinct from fiscal policy.1 It advocates an independent central bank which sets interest rates to target inflation. It also assumes a binding solvency constraint on fiscal policy, which in its strongest form is the “sound finance” condition discussed above. Frequently , mainstream macro relaxes the constraint to say that debt cannot breach some debt-to-gdp ratio, or must remain stable. The existence of this solvency constraint motivates the creation of an independent central bank. The budget deficit(surplus) alone cannot target inflation and keep the government solvent in the traditional sense, or maintain a stable debt-to-gdp ratio. In that case, there are two targets, but only one policy instrument,2 and it is thus impossible for the single instrument to successfully hit both targets.

By adding a second policy instrument, and assigning it to the output gap/inflation, the standard macro set up allows the budget balance to maintain government solvency or satisfy the debt-to-gdp boundedness constraint.3

MMT

So in an MMT world/model, fiscal policy does all the work in targeting inflation and closing the output gap4, and can do so because the nominal budget constraint on the government is removed. There is no need for monetary policy independent of fiscal policy. Many proponents of MMT suggest that short term interest rates should be kept permanently at zero, but the exact level actually should not matter. The government has no use for a second instrument because it only has one target to hit.

MMT does not say that government spending should be unlimited. It says that the government should run a deficit/surplus such that the output gap is as close to zero as possible and thus inflation is low and stable. “Doing MMT” does not make it easier to pay for big ticket items like Medicare for All. If such programs generated excessive deficits, they would be inflationary. MMT does not say taxes are unnecessary. Taxes are still needed to keep deficits in check so that they don’t cause excessive inflation. All MMT actually says is that since the government needn’t worry about a nominal solvency constraint on its own spending, it is sufficient to use fiscal policy alone to target inflation or the output gap. (See the footnote for a more precise explanation.) The biggest insight that MMT does have is that if a large shock hits the economy, fiscal policy makers should not hesitate to act decisively with large deficits to blunt it. In my view, this lesson helped avoid a global depression during COVID but policy makers may have overshot the runway and caused inflation by maintaining large deficits for too long after the economy reopened.

Conclusion

MMT fits comfortably inside standard macro. It is not a claim that deficits do not matter. It is a claim about which constraint binds. For a government that issues its own free-floating currency, the binding constraint is not nominal solvency but real capacity. The government can always create the money needed to meet nominal obligations, but it cannot create labor, housing, energy, medical care, or productive capacity by keystroke. Past full employment, additional nominal demand becomes inflationary.

Viewed this way, MMT is not a radical break from mainstream economics, but rather an alternative assignment of policy instruments. Once the nominal solvency constraint is removed, fiscal policy alone can be used to stabilize inflation and the output gap, while taxes remain essential for managing aggregate demand and containing inflationary pressures. Whether one ultimately agrees with MMT or not, the theory is best understood as a debate about the nature of government constraints and the appropriate role of fiscal policy, rather than a claim that governments can spend without limit or consequence.

Asset Market Equilibrium

So much of the housing discourse becomes unmoored from reality because people do not view all real estate as an asset class with a rate of return. My work has touched on this key truth time and time again. Any asset can be priced as just the discounted stream of revenue it generates. When platforms like AirBnB or VRBO reduce the frictions in converting housing units into de facto hotels, they effectively increase the revenue stream a property can generate, with minimal additional conversion costs from the property owner. This must cause the asset to reprice. Or in other words, in increases home values.

On the rental side, with the advent of platforms to market and rent out homes as STRs, landlords (and even homeowners) now have the choice between offering their properties as long-term or short-term rentals. The proportion of short-term to long-term rentals is determined by the asset market equilibrium. When risk-adjusted returns to short-term rentals are higher than long-term rentals, it encourages conversion. This removes housing from the long-term rental market and pushes up rents, until the marginal landlord is indifferent between using their property as a short or long-term rental.

Empirical Findings

The theoretical mechanism is simple and easy to understand. However, does it hold up empirically? Yes. Baron et al studied data from AirBnB listings across the US. They found a 1 percent increase in listings led to a 0.018 percent increase in rents and a 0.026 percent increase in home values, with much larger effects in tourist-heavy areas. Another study in Barcelona found that in tourist-heavy districts with the most AirBnB activity, AirBnB listings drove up rents by 7 percent and property values by 17 percent. A third study focused on Boston showed that an that an increase in AirBnB listings modestly raised asking rents, and reduced the availability of long term rentals.

In any event, though I am sure the econometrics nerds can point out issues in all of these studies, the fact that they largely confirm theory and mostly reach the same conclusion, that the effects of AirBnB listings on home prices and rents are modest but real, should put a lot of the doubt to rest.

Winners and Losers

Market forces are constantly producing winners and losers. These are not economic externalities per se, and frequently, arguments come down to equity, or the distribution of welfare, not efficiency, which is about maximizing welfare regardless of distribution. There in no doubt that AirBnB is economically efficient. The losses suffered by tenants are perfectly offset by the gains to landlords, and AirBnB produces another set of winners, travelers who get more lodging options at cheaper prices. Other winners may include people whose incomes depend on tourism related industries like dining and entertainment. But the equity question cannot be dismissed. The people negatively impacted by short-term rentals tend to be the poorest and most financially vulnerable. In the end, this issue is just one more example of how most housing issues just get back to inequality.

Conclusion

The economics of short-term rentals are neither mysterious nor especially controversial. When a technological platform increases the revenue potential of housing units and lowers the friction of converting them into tourist accommodations, housing prices should rise. At the same time, shifting units from the long-term rental market into short-term use reduces the stock of rental housing and pushes rents upward. This is exactly what standard asset pricing theory predicts, and it is broadly what the empirical literature finds. The magnitudes may vary across markets, and AirBnB is clearly not the sole driver of the post-2019 housing cost increase, but the direction of the effect is difficult to dispute.

The real debate is therefore not about whether short-term rentals affect housing markets, but about how policy makers should balance competing interests. Short-term rentals create substantial benefits for property owners, tourists, and local businesses tied to the visitor economy. Yet those gains often come at the expense of renters and lower-income residents in high-demand neighborhoods, who bear the costs of higher housing prices and reduced availability. In that sense, the conflict over AirBnB is ultimately part of a broader political and economic question: who benefits from rising asset values, and who is priced out by them.

The PE Hack Job

Private Equity operates in many different ways, but the most infamous is what I will term the PE Hack Job. A private equity firm purchases a company, sells off its assets, and shuts it down, hopefully at a profit. This of course displaces many workers and puts a company which may have been financially stable, in sense that it made consistent accounting profits, out of business. The PE hack job is fundamentally profitable because the returns generated by the firm are well below what could be generated by selling off its assets and investing the proceeds elsewhere. In other words, a firm does not cover its cost of capital.

A Paradox

The common thread with PE is cost of capital. PE loves to swoop in and liquidate companies with valuable assets who do not cover their cost of capital. However, the reality, as revealed by economic research, is that most firms, even publicly traded ones, fail to cover their cost of capital, that is generate economic profits. They of course generate accounting profits and thus might never be forced to shut down, but the profits are consistently below an adequate return given risk. This reality is consistent with the fact that in large stock market indices, a small minority of companies generate most of the growth.

Why do firms which systematically fail to clear hurdle rates remain open? Original owners are loathe to voluntarily shut down for a variety of reasons. First, being in business is a fundamentally an optimistic activity. Even if profits are low now, most business owners are optimistic about their ability to grow profits in the future, even if cold, dispassionate analysis indicates otherwise. Second, many business owners feel a loyalty to employees, and try to stay open as long as possible even if things look grim. Therefore, a firm which experiences economic losses, that is, profits below what could be generated by liquidating the company’s assets and investing the proceeds elsewhere, usually do not shut down until they actually start suffering accounting losses. Private equity funds have no such loyalty to other stakeholders such as employees. As such, they will buy a company and shut it down if they think they can profit from the sale of the assets.

However, this presents a strange paradox. If all companies which did not cover their cost of capital shut down, perhaps 60 percent or more of businesses would close their doors. This would put over half the labor force out of work, cause an economic depression of unimaginable scale, and ironically, likely cause many firms which previously did cover their cost of capital to go out of business for lack of customers. Or put in a different light, basic economics professes that welfare (and growth) are maximized when capital is channeled into the most efficient outlet. However, the macro-economy depends on “laggard firms” staying in business, since most excess profits are generated by a select group of firms, which do not employ very many workers as a share of the overall labor force, but who depend of workers of less efficient firms as customers.

Efficient at the Firm Level?

But is PE even efficient at the firm level? One argument is that PE is economically efficient in that it channels capital (and by extension labor) into more profitable outlets. A legacy firm which does not cover its cost of capital pulls labor and inputs away from more efficient firms. Keeping this businesses open ought to slow economic growth. After all, letting them fail should redirect workers to more productive firms. Unfortunately, this thinking assumes away considerable frictions in the labor market. Workers who become unemployed typically spend three to six months looking for a new job before obtaining one, and often times over a year if the job market is weak. Once hired, workers usually spend a significant amount of time onboarding, during which time they are learning the new job and their skills are not fully utilized. So the trade off is not truly between employing workers at a low productivity firm or high productivity firm. When workers lose their jobs, they frequently don’t work at all for long stretches. This income is lost forever, as is the external surplus that this labor would have generated both for shareholders and customers. Thus, when accounting for labor market frictions, it is unclear if PE is welfare maximizing (across all stakeholders) and therefore efficient.

Conclusion

The private equity business model starts from a premise which is widely accepted in economic orthodoxy. Capital should be channeled into the most efficient mechanism. However, in the real world, only a small group of elite firms actually generate economic profits, the rest only generate accounting profits. In theory, these firms should shut down and invest their capital in higher returning vehicles. But if practiced at scale, the economy wide disruption of laying off over half the labor force would like cause even “elite” firms to fail. This creates a strange paradox At the firm level, PE just efficiency channels capital into higher earning outlets, but if done at scale the massive labor market disruption would implode the economy. Even at the firm level, PE might not even be welfare maximizing (again, in aggregate since workers always lose). The channeling of a “laggard” firms’ capital into more profitable outlets certainly maximizes shareholder value. However, since displaced workers often spend months or even over a year finding new employment, the income and external surplus that would have been generated by those workers is lost forever. Once accounting for these losses, the efficiency gains from private equity may well be fully cancelled out.

Single Family Homes are Usually Bad Rentals

While there appears to be a political consensus that institutional ownership of single family homes is undesirable, the reality is that most single family homes are not under “threat” of being purchased by Wall Street. The simple fact is that rental yields are not high enough to clear hurdle rates or in many cases even cover interest payments on debt used to purchase the home. In most high cost or medium cost markets, these conditions prevail and it is enough to keep big investors away. Where investors have been active is in low cost and highly affordable markets. Think Tulsa, Oklahoma or Jackson, Mississippi. In these markets home prices are still staggering low, and thus rents do not need to be very high in order for the property to generate a high rental yield. Institutional ownership in high cost markets is virtually non-existent and many landlords are small time “accidentals” who needed to move quickly and did not have time to sell their house and rented it instead. The “golden age” of single family rentals was really in 2010s, after home prices crashed but rents surprising held up even during the Great Recession, perhaps because so many families lost their homes to foreclosure and became renter households.

The effects of institutional ownership on the single family home market is not necessarily inherently undesirable. When rents for single family homes are high relative to prices, that is, rental yields are high, it signals that there is a not enough rental housing relative to housing which is for sale. Allowing some homes to purchased by investors and converted to rentals will raise prices but lower rents as new rental properties hit the market. This of course benefits renters at the expense of home buyers, but it is just the housing market reaching a new equilibrium, where more rental housing gets supplied to the market because that is where the need is.

In general however, it is important to note that in most markets at most times, the highest best use of a single family home is as an owner occupied dwelling. So how does build-to-rent even make sense economically? That’s what the rest of this article is about.

Build-to-Rent

Single family homes typically make very unattractive rental properties. It therefore may come as a surprise that building single family homes with the intention to rent them out could ever make sense economically. However, this is where design comes in. Unlike traditional single family homes, these units tend to be smaller, and on smaller lot sizes. Mechanically, this allows properties to generate more rent per square foot than a typical single family dwelling, significantly increasing returns. It is also important to note that only 20 percent of build-to-rent single family homes are even fully detached. For this reason, this building type is sometimes called “horizontal multifamily.”

At the end of the day, any build-to-rent community is designed significantly different from a development intended for owner occupants. This enables projects to deliver a high enough return on investment to justify their construction.

Conclusion

Single family rentals are usually poor rentals because they do not generate high enough rental yields. For this reason, the overwhelming majority of newly constructed single family homes are built in order to be sold to owner occupants. In rare cases, usually in highly depressed/affordable markets, single family homes can occasionally be an attractive investment as a rental property. This is because homes are so low priced that the rents they generate provide an attractive yield.

Build-to-rent bucks this trend by designing single family units with renters in mind. Units tend to be much smaller than owner occupied single family dwellings. Fundamentally, its a different product type built for a specific kind of customer. This enables the numbers to work and for projects to clear hurdle rates, overcoming the challenge that most single family homes face when used as rental properties.

The United States faces a malaise where a majority of citizens believe the country is on the wrong track. This phenomenon has shown up consistently in polling since the early 2000s. There has also been a precipitous drop in social trust, and a general lack of faith in institutions, and the government’s ability to solve problems. This belief is false. The US government has an incredible track record of success in confronting historic challenges. But here’s the rub, in order to solve large problems, there have to be large problems in the first place. I am all for innovation and progress, but the historical reality is that crises and conflict are the ripest time for major actions by governments. While the political unity of theses crises allowed for large scale government actions which for the most part succeeded, the crises themselves left lasting scars that no amount of problem solving could address.

Lincoln Did Much More Than Win the Civil War and End Slavery

Abraham Lincoln’s greatest accomplishments were winning the Civil War and ending slavery, but by any measure the Lincoln administration created the modern Federal Government. Lincoln introduced the country’s first paper money, the nation’s first income tax, passed the National Bank Act which further standardized bank notes across the country and was a precursor to the Federal Reserve System, created a national system of universities with the Morrill Land Grant Act. Lincoln also passed legislation to fund the Transcontinental Railroad, one of the biggest engineering projects of its time. Lincoln was able to pass all of these measures because his political opposition was actively trying to secede from the Union and thus did not send Representatives or Senators to Congress. There is no doubt that the costs of a bloody civil war, 600,000 Americans dead, the South left in ruins were worth it to end the evil of slavery. However, the Lincoln administration did not let the crisis go to waste. They used the temporary absence of any significant political opposition to pass a program of sweeping change which still shapes American life today. Again, the Civil War was worth it to end slavery. But was the Civil War worth it to get the National Bank Act passed? Very few who lived through it would answer in the affirmative.

The Manhattan Project

The Manhattan Project was the largest scientific project ever undertaken by the US government and perhaps in history. It brought together thousands of scientists across several top secret sites throughout the country, including the now famous Los Alamos National Labs in New Mexico. At its peak, the project employed close to 130,000 thousand workers. The project not only developed nuclear weapons, but it made many breakthroughs in nuclear physics which later led to civilian nuclear power. However, despite its smashing success, we cannot ignore that the catalyst for the project was fear of Germany developing nuclear weapons first, as well as a desire to win the war generally. The mass mobilization of scientists and engineers was born out of a worldwide war. Was it worth it win WWII to defeat the Axis powers and the unspeakable evil those regimes represented. Yes. Was the war worth it so that we got nuclear power? Doubtful.

Operation Warp Speed

The most recent large scale science project of the US government was Operation Warp Speed. The COVID 19 pandemic created a national crisis and the development of a safe and effective vaccine became the holy grail for policy makers. Interestingly the “innovative” part of the project was as much economic and biological. Effectively, the government agreed to foot the bill for development and absorb the risk of vaccine failure by agreeing to per-purchase millions of doses of vaccines. The upside of vaccine success however was largely left in private hands. This created a humongous incentive for the pharmaceutical companies to devote resources to vaccine development. The result was a vaccine in under year, when the typical timeline for drug development is 15 years. A splendid and amazing technological achievement which saved millions of lives and also gave birth to mRNA technology. But yet again, do we want another public health crisis like a global pandemic to spur on medical research? Who wants to go back into quarantine?

Conclusion

Necessity is the mother of invention, and the US government has a strong track record of rising to occasion to solve big problems with technology or new institutions when history demands it. Nor are these feats of wonder in the distant past. As recently as 2020, a US government program, Operation Warp Speed, saved millions of lives globally. That said, the crises which created the urgency to develop these innovations brought on tremendous costs we are still feeling today, especially in the case of COVID 19 pandemic. I am thankful that our government was able to stand and deliver when it needed to in the past. But I truly wish that such crises become less frequent in the future. Having fewer big problems will always beat strong problem solving capacity.

LIHTC

A full overview of the ins and outs of the LIHTC is well beyond the scope of this article. Books could be written on the subject. However, to proceed and to understand the significance of the recent change, we need a working knowledge of the LIHTC. I have decided to add in some interesting details about the program that are not strictly necessary to understanding the policy change at issue, but which are interesting themselves and some readers with an interest in housing finance might find useful. Readers familiar with LIHTC may wish to skip this section.

LIHTC comes in two main forms, 9 percent tax credits and 4 percent tax credits. These percentages correspond to the so-called qualified basis, roughly speaking total project costs excluding some important items such as land. The tax credits are good for ten years. Say a project has a qualified basis of 10 million dollars. If it were a 9 percent deal, it would generate an annual tax credit of 900 thousand dollars, and if it were 4 percent deal, it would generate an annual tax credit of 400 thousand dollars. These tax credits are sold off to investors to raise equity to fund affordable housing construction. The tax credits, along with depreciation, tend to generate a tax “yield” of about 4-6 percent annually. The project itself typically generates little to no residual value to investors, since rents are capped and must comply with affordability rules.

Astute readers will notice that a 4-6 percent return is quite paltry in the real estate space, and is substantially lower than what market rate projects typically must return to be attractive to investors. This is because LIHTC deals face a different set of risks and generate value for investors via tax savings, which is guaranteed. (More on this in a moment) The tax credits do not depend on the financial performance of the project. They are solely determined by the cost to build it.

Contrast this with the risks faced by investors in market rate projects, who derive most of their returns from capital gains driven by steady rent increases. These rent increases may not materialize if the rental market softens and are far from a sure thing. Market rate projects present far more financial risk to investors, and thus require higher returns. On the other hand, LIHTC deals resemble tax advantage fixed income far more than real estate equity. The returns come from tax savings, and are not tied to the capital gains or the financial performance of the project. The real risk LIHTEC investors face is that projects do not comply with affordability rules, in which case the tax credits can be recaptured by the IRS.

The Gating Mechanism

Nine percent LIHTC tax credits are limited by statute and are allocated annually to state housing agencies. Four percent credits are not, but they are effectively limited by another key requirement. For a project to qualify for 4 percent LIHTC credits, 50 percent of the total costs must be financed with private activity bonds, or PABs. These securities are tax-exempt bonds used to finance private activity which is in the public interest, like affordable housing. And here is where the 4 percent credits are capped in practice. Each state is given an quota for PAB issuance based on population. Thus, since there is a maximum volume of PABs allowed to be issued in each state annually, they act as a key limiting factor on the flow of 4 percent LIHTC deals. Without this indirect limit, a theoretically unlimited amount of tax credits could be generated via 4 percent LIHTC. Congress wished to avoid this, since the credits deprive the government of revenue, and thus created a strange mechanism to cap LIHTC deal volume in each state.

A Lowered Threshold

Beginning this year, the required threshold of PAB financing to qualify for 4 percent LIHTC tax credits is 25 percent. This greatly lifts the ceiling on the total number of LIHTC projects which can proceed, since each individual project will consume less of each state’s PAB issuance cap.

The change also, perhaps counter-intuitively, reduces construction risk. A project which raises exactly 50 percent of its total costs in PABs but suffers a cost overrun will not comply the rule and will lose access to 4 percent LIHTC credits if non-PAB financing sources are used to fill the gap. Thus, LIHTC developers tend to target at least 55 percent PAB financing in their capital stacks to create a cushion. With a lower threshold, it is less likely that a cost overrun will cause a project to run afoul of the rules.

Finally, it is important to remember that this change is not without fiscal costs. By allowing more LIHTC credits to be created, the government deprives itself of tax revenue. LIHTC is a very real subsidy for affordable housing.

Conclusion

LIHTEC is a complex financing mechanism to fund affordable housing. While attention is currently on the ROAD Act, lowering the PAB funding threshold to qualify for 4 percent LIHTC credits greatly lifts the indirect cap on the total number of LIHTC deals which can go forward in a given year. It also has the added benefit of reducing construction risk. Though not without fiscal costs, this change has the potential to meaningfully spur a substantial increase in the construction of affordable housing.

The Passage of PROMESA

After years of unsustainable borrowing and overly optimistic budget projections, Puerto Rico entered a fiscal crisis in 2015. The island and its instrumentalities had accrued 72 billion in debt in the form of various types of securities. In comparison, the value of the entire island economy was at the time roughly 70 billion dollars annually. On June 28th of that year, then Governor Alejandro Garcia declared Puerto Rico’s debts “unpayable.” On May 1st 2016, Puerto Rico unilaterally suspended debt service payments. On June 9th 2016, Congress passed the Puerto Rico Oversight Management and Economic Stabilization Act (PROMESA) which among other things, 1) Provided a legal mechanism by which Puerto Rico could suspend debt payments and enter a bankruptcy process. 2) By virtue of (1) shielded Puerto Rico from litigation stemming from non-payment 3) Established an oversight board to approve Puerto RIco’s annual budget and negotiate on its behalf with creditors.

The last of these provisions is by far the most controversial, but before diving into the now infamous board, it is crucial to examine the legislative history of PROMESA. PROMESA was passed on a broad bipartisan basis and after extensive lobbying efforts by elected officials from Puerto Rico who were members of both major island political parties. These officials included both the Governor and the Resident Commissioner, the island’s non-voting member of the House of Representatives. Both men are members of different political parties and represent opposing sectors of the Puerto Rican electorate. Therefore, broad consensus existed both in Congress and among Puerto Rico’s elected officials that a bill along the lines of PROMESA needed to be passed.

The Original Opposition to PROMESA

Opposition to the bill came from two places. First, conservative Republicans opposed any legislation that would allow the Puerto Rican government to discharge debt via bankruptcy. This was exactly the reasoning of Jenniffer Gonzalez, then minority leader of the Puerto Rico House of Representatives and now Governor when she voiced her opposition to any bankruptcy bill in an OP ED. Although mainland party affiliations and right/left divides are not important in Puerto Rican party politics, Gonzalez is a Republican, is deeply economically conservative in her personal capacity and caucused with Republicans in Congress when she was Resident Commissioner from 2016 to 2024.

Another major politician to oppose PROMESA was Ricardo “Ricky” Roselló, son of former governor Pedro Roselló, and who at the time was running for governor against Pedro Pierluisi in pro-statehood party (PNP) primary. While Roselló may have sincerely believed that Puerto Rico could pay all its debts without restructuring, his campaign was also financed by bondholders who opposed PROMESA for obvious reasons.

Opposition also came from the progressive wing of the Democratic party who favored direct cash transfers to the Puerto Rican government in order to avoid cuts to government spending at the local level. A direct cash bailout for Puerto Rico, while perhaps desirable given the economic chaos on the island, was a non-starter for the Republican controlled Congress at the time. Thus a bipartisan, centrist compromise emerged which was ultimately passed by Congress and signed into law by President Obama.

The Rewriting of History

Since 2016, many have tried to rewrite the legislative history of PROMESA. A highly vocal revisionist school has emerged, claiming that PROMESA was extremist, anti-Puerto Rico legislation passed over the voices of Puerto Ricans. This school also claims that the central purpose of PROMESA was to protect Puerto Rico’s creditors rather than to provide a mechanism for Puerto Rico to seek bankruptcy protection and debt restructuring. But this is convincingly disproven by both the lobbying activity of Puerto Rican elected officials from both major parties, and by the votes of Members of Congress themselves. Indeed, the most extreme pro-creditor position, that Puerto Rico ought not to be able to discharge any of its debt, came not from the bill’s proponents but from its detractors. Indeed, of the 127 “no” votes, there were 24 progressive Democrats and 103 pro-creditor Republicans.

Others contend that the ability of the oversight board to veto local laws is rank colonialism. This claim ignores key context. In practice, virtually all of the laws vetoed by the board have been cases where the government passed laws which spent money without raising equivalent revenue. While it some may contend that enforced budget discipline is paternalistic colonialism at its worse, the reality is that Puerto Rico lacked and lacks access to willing lenders to finance deficit spending AND such spending would run afoul of the procedure of federal bankruptcy courts. Allowing debtors to incur more debt while the resolution of existing claims of previous creditors are still being negotiated would create chaos and it is thus no surprise that bankruptcy courts prohibit it. In sum, the power the board exercised over Puerto Rico’s budget is largely redundant to what any bankruptcy court would have exercised, and necessarily so to ensure an orderly restructuring of Puerto Rico’s debts.

Finally, the authority the board has over Puerto Rico’s laws is not unlimited. PROMESA states in section 108, article A, that:

[Puerto Rico may not ] enact, implement, or enforce any statute, resolution, policy, or rule that would impair or defeat the purposes of this Act, as determined by the Oversight Board.

Which means that Puerto Rico is largely prohibited from passing legislation that would sabotage the restructuring process. This preemption of local law is not without precedent. In fact, section 1123 of the Federal Bankruptcy code states that bankruptcy courts can:

1. specify the treatment of any class of claims or interests that is impaired

2. provide adequate means for the implementation [of the restructuring of those claims]

And crucially do so:

Notwithstanding any otherwise applicable nonbankruptcy law

While US States currently do not have access to bankruptcy, it would simply be unfathomable for Congress to create any form of bankruptcy for states which did not include similar language. To do so would allow states to rewrite the rules immediately before entering bankruptcy, most likely by elevating claims of politically favored creditors, or to evade creditors altogether by simply passing budgets which did not appropriate funds to make payments toward the final plan of adjustment ratified by the bankruptcy court. That is, “provide adequate means” entails that the bankruptcy court could order the bankrupt state not to enact budgets which renders the final plan of adjustment of its debts financially nonviable, and crucially, the court can do so even over the objections of the elected government of the state. That said, neither the board or bankruptcy courts can veto laws which don’t affect the restructuring process or its financial viability. Again, their powers are not without limit.

None of this is to say that bankruptcies are not anti-democratic in nature when the debtor is a polity. However, the anti-democratic nature of public bankruptcies is a feature and not a bug. Bankruptcy courts have to examine the legal validity of claims against the debtor, and weigh fundamental rights of property against economic and financial realities. Such a task is wisely put in the hands of courts, not of elected officials. Such was the case in Puerto Rico, and it is not speculative to say that it would almost certainly be the case if a state were to be given access to bankruptcy.

The Results

But what did/has PROMESA accomplished? Before PROMESA was passed, roughly one fourth of Puerto Rico’s annual budget, or 4 billion dollars, was dedicated solely to debt service payments. For a brief time, Puerto Rico was able to cover these payments with even more borrowing, but in the years leading up to the crisis, the government inevitably had to cut back on essential services to keep its head above water. This dynamic was exacerbated by the Great Recession, which hit the island particularly hard. By initiating a bankruptcy process to restructure Puerto Rico’s debts, PROMESA allowed Puerto Rico to legally halt debt services payments while the bankruptcy case was being litigated. Since the final agreement was not ratified until 2022, Puerto Rico was able to legally evade its creditors for nearly six years! Contrary to popular the belief, most of the savings went back into essential services like schools, police, and salaries for chronically underpaid public employees.

When an agreement was finally reached with Puerto Rico’s creditors in 2022, roughly half of Puerto Rico’s 72 billion dollars in debt was erased, and crucially, debt service payments were capped a roughly one billion dollars a year to prevent them from crowding out essential services in the government’s budget. There is no way to spin it. This is a massive win for Puerto Rico and massive loss for bondholders.

PROMESA’s unfinished business continues to be the restructuring of roughly nine billion dollars of debt tied to Puerto Rico’s public power company, known as PREPA. Creditors have continued to resist a consensual agreement like the one reached with the government itself. However, the benefit of endless delay has largely accrued to Puerto Rico’s ratepayers, who would need to pay substantially more for electricity in PREPA still needed to service legacy debt. Just like the government’s debt, any payments to bondholders remain frozen until a final agreement is reached.

Conclusion

PROMESA was a bipartisan compromise to address Puerto Rico’s debt crisis. It is wrong to label the law as anti-democratic. The board’s powers are limited to budget matters. And more importantly, litigation to which the government is a party is inherently anti-democratic by design, because fundamental issues of law and rights must be adjudicated. It would be absurd to allow the government’s elected officials to simply pick and choose which debts to pay. That would make a mockery of property rights. But biggest indicator of PROMESA’s success is the results. Pensioners did not see a dime cut from their benefits, half of the island’s debt was wiped out, public services were maintained or in some cases expanded, and debt service payments are capped below a manageable level. Let’s not rewrite history. PROMESA worked.

What’s a Home Price Index?

In the broadest possible sense, a home price index is an intended measurement of the overall value of the housing stock. Clearly, as more units of housing are added to the housing stock, the value of the aggregate stock of housing must increase. Therefore, the aggregate number tells us very little about how much individual homes are appreciating (or depreciating) in value. Therefore, virtually all home price indices try to capture some average or median home value, which by definition adjusts for the number of housing units. This approach sounds simple but it actually presents some annoying problems.

All Transaction Indices

We actually don’t know the exact market value of any home until it actually sells. No two homes are alike, and every unit of housing has a unique location. One popular approach to creating a home price index may be track the prices of all home sales over a given time interval and report the median or mean. This method is known as a all-transaction index. However, the homes actually put up for sale is not a random sample of all homes, so trying to extrapolate a population mean or median from it will entail bias.

In general, the bias is upwards, because higher-end homes tend to transact more often. Say a neighborhood has 10 homes. Eight of them are modest three-bed, two-bath homes, but two are significantly larger. If during a given period both large homes sell but the other homes do not trade, an all-transaction index would grossly overestimate home values for this hypothetical area.

Another bias created by an all-transaction index is that home sellers may be reluctant to sell their homes if they have doubts about the condition. Or similarly, many home sellers improve their homes in anticipation of a sale in order to fetch more money. This also creates upward bias in an all transaction index, because homes in worse condition which never go on the market are excluded.

Lastly, all transaction indices must contend with the bifurcation of the new versus existing home market. Fundamentally, newly constructed homes and existing homes are different product types and different statistical populations. The composition of new builds in terms of size and quality can change rapidly. Imagine a period where many condominiums were built followed a period where most construction was single family homes. Here, a index of new home sales is mostly measuring the composition in terms of size and quality of the homes being added to the housing stock, not an increase in home values per se. For this reason, most indices differentiate between new and existing home sales.

Repeat Sales Index

To address these biases and compositional effects, many indices use the repeat sales method. Rather than measure the median price of all homes that transact, the repeat sales index looks at the change in price since the last time a home sold. This controls for home size or quality, since the same homes are compared to themselves.

There are two main disadvantages of repeat sales indices. First, they are costlier and require more data to construct, and thus only exist for certain areas. Second, they trade one form of bias (composition) for another. Repeat sales indices must use homes that have transacted at least twice. This by itself is not an insurmountable problem since sales records often go back decades. However, the index must adjust for the time between sales, and most come up with a weighting system to determine how time between sales determines the impact on the overall index of a given data point. The intuition here is that two relatively close sales are likely to have data which is less “noisy,” and long time intervals between sales increase the probability that substantial improvements have been made to the property which affect value. The mathematics of the weighting system to account for time between sales is beyond the scope of this article, but it is not a perfect process and does not strip out all bias in the index.

Conclusion

Home price indices are useful tools but are not without biases. All transaction indices suffer from compositional effects because more valuable homes tend to sell more often. Repeat sales indices attempt to fix this issues by comparing homes with themselves over long time intervals. However, they are more expensive to construct and require more data. Lastly, the weighting system with regard to time between sales itself is imperfect and may create additional biases.