Taxing rezoning windfalls (betterment)

This is an excerpt from a new paper of mine available here. 

What is the source of rezoning gains?

The answer to this question relies on understanding the nature and extent of property rights. A core right of a property owner is to exclude others. A right holder of this type usually also has a residual claim to income generated in the space over which they have exclusionary rights (“pay to use this space or I’ll exclude you from it”).

But these rights are limited. They evolve as the law evolves, including zoning and planning law. Rights to access and sell minerals within a property’s exclusive space, for example, are not part of the private property rights bundle in Australia. They are instead owned collectively by States.

The value of private property rights is primarily determined by how much revenue can be generated above costs (excluding property rights costs) on a site at a particular location. The value of property rights is a residual, just as other assets like shares in company ownership represent residual claims on income. Whichever use of a site generates the largest residual income is its highest value use. 

The revenue potential of private property rights is affected by market conditions, location-specific features, fixed improvements, and the nature and extent of the property rights bundle itself. Investing in new fixed improvements, like buildings and earthworks, can add to property value because they remove a cost to gaining revenue (remember, property value is revenue minus costs needed to earn that revenue at that location). The value of fixed improvements becomes “attached” to the property as they cannot be separated physically or legally from the property right to a location.

However, the value of private property rights can also increase due to external factors that are not due to investments made by the property rights owner. When this occurs, it is known as betterment. For example, new public works that improve the accessibility of a location will increase the value of those sites because they reduce the transport costs of using those sites to generate revenue. Similarly, when the law changes the nature or extent of property rights in a way that increases the revenue-generating potential of that site, such as with rezoning, the resulting increase in value of the property rights bundle is betterment. 

Planning rules coordinate property uses across locations by defining property rights bundles. For this coordination role to affect property uses it must legally restrict some uses so that the highest value legal use is different to what might occur with a “no zoning” property rights bundle. Hence, there is an inherent conflict between maximising the value of any individual property and coordinating the location of property uses across a region. 

The diagram in Figure 1 shows how betterment arises conceptually and how a tax on betterment transfers value to the public that would otherwise accrue to private property owners. 

Figure 1: Betterment in the residual model of property value

The left column shows the value of the property rights at a site, such as agricultural or industrial land. That value is determined by the revenue that can be made from exclusively using that site, minus the necessary costs. The site’s property rights value is the residual. 

Suppose the market value of the output being produced increases, such as the value of crops on an agricultural property, but all input costs remain the same. In that case, the site value rises to reflect the higher residual value of production on that site. The new site value and its betterment component from these external factors is shown in the second column of Figure 1. Another example is if the rent of a residential dwelling rises, but the costs of operating that dwelling remain constant. This increases the property value. Where a property tax system exists, some of the value gains accruing to property rights from higher market prices of production are shared with the public.

The third column of Figure 1 shows what happens if there is a change in the nature or extent of the property rights through rezoning. For example, if the previous highest value use of a site was industrial only because zoning laws prevented other higher value uses such as residential, then rezoning will change the highest value legal use of the site and hence its value. The difference between the “before” site value (V1) and the new “after” site value (V2) is betterment. 

A flexible planning system presents opportunities to increase revenue from development by exceeding codified density limits and hence offers another way to generate betterment for a landowner. The betterment gained from exceeding codified planning density limits is just as real as the betterment gained from rezoning. It is why, despite cost, risk, and time involved, many property owners seek planning dispensation instead of complying with codes and gaining fast approvals. 

Here is a useful way to conceptualist betterment. Rezoning adds an additional private property right to the previously owned bundle. The value of this new right is betterment, and it reflects what the market would pay if these new property rights were instead auctioned for sale to existing property owners.

The effect of a tax on betterment arising from rezoning is illustrated in the fourth column of Figure 1 using the 50% tax rate proposed in Victoria. A share of the betterment value is transferred from the private property owner to the public, reducing the private payoff from rezoning decisions and increasing the public’s share. 

The size of betterment from rezoning can be extraordinarily large, often many multiples of the previous site value. For example, a well-situated industrial site in Sydney’s inner west was bought for $8.5 million, rezoned high-density residential, then sold again just a few years later for $48.5 million.  

Rezoning at Fishermans Bend in Melbourne led to numerous sites trading at values many multiples of their previous industrial use values. At 320 Plummer Street, Port Melbourne, the 7,468m2 site formerly used as the Rootes (Chrysler) factory transacted in 2009 for $1.7 million with industrial zoning rights. After rezoning to high-density residential, an application was made and approved for the development of three towers containing 443 dwellings and 908m2 of commercial floor area. The property was subsequently purchased for over $11 million. Even after inflating the 2009 sale price to $3 million to reflect sale prices in 2015 for similar industrial properties, the windfall rezoning betterment is roughly $8 million, with the site trading for nearly four times its previous value.  

Notably, betterment is not an additional cost to housing development. The value of property rights is a residual; after all, there are no input costs to these rights. Betterment merely reflects the change in the market value of the residual claim on income of the property rights owner, and taxing betterment merely transfers this value from the property rights owner to the community. 

Harvesting housing supply

There is a common thread between the boom in lumber prices in the U.S. and dynamic models of housing supply. 

That common element is that lower interest rates make it optimal to both harvest trees slower and to build new homes slower. Let me explain.

Take a look at the diagram below representing a forest with trees at different stages of growth.

The stock of potential timber from trees at different stages of growth is marked as Q. So 10 tonnes of large trees, 7 tonnes or medium trees, and 3 tonnes of small young trees. A total of 20 tonnes of timber there. 

But each stand of trees with a different maturity has a different rate it grows per year. Young trees accumulate timber faster (proportionally) than older trees. The growth rates for the trees in each stand are marked—the oldest grow at near enough to 0%, the next at 5%, and the youngest at 10% per year.  

Given this forest with a stock of 20 tonnes of timber, how much is optimal to harvest each year?

The trick to thinking about this question is to realise that the trees are assets, quite literally growing in value each year. When you harvest a tree you are swapping a "tree asset" for a "cash asset". The way to tell what is optimal is to compare the return from the tree asset that you give up to get the return from the cash asset.  For example, if you can earn a 7% return from cash, you won't harvest a tree that is growing in size, and hence value, by 10% per year. It's a losing trade to give up 10% to get 7%. 

In this forest, we harvest 10 tonnes this year if the interest rate is below 5%, and 17 tonnes if it is above 5%. 

When it comes to forest management, lower interest rates mean slower harvesting of timber. If interest rates fall from 7% to 3%, then all the trees growing at a rate between 3% and 7% per year should be left to grow rather than be harvested and replaced with saplings. This is well understood when it comes to harvesting forests. 

But it is not well understood when it comes to "harvesting housing development opportunities". 

Undeveloped urban land is a lot like a tree—it grows in value without being developed (harvested). Like our forest, we don't just develop all land as soon as the revenue exceeds the cost. We optimise the rate per period to maximise value from the site (or set of sites). This is why developers stage housing subdivisions as much as possible. 

A lower interest rate changes the trade-off between owning undeveloped land and getting cash from development. It makes a slower pace of housing development optimal, all else equal. 

Not all else is equal, obviously. The price adjustment to lower interest rates generates demand that new housing supply responds to—interest rates are not the only factor. In the last 20 years we have relied on this temporary demand-boosting effect of interest rate reductions to generate supply, but this has led to structural low-interest-rate conditions that will not encourage supply when demand falls. 

In housing, optimal harvesting is called the "housing supply absorption rate". I explain it here. 

Stamp duty for land value tax

The NSW government is proposing to give homebuyers the option to not pay stamp duty on their housing purchase and instead opt to pay an ongoing land value tax. I have labelled such policies SD4LVT.

SD4LVT seems to be motivated by "bad economics". All the efficiency gains that economic analysts claim will occur are merely assumptions and not very realistic ones at that. 

A major consideration for SD4LVT is its effect on housing prices. While SD4LVT is often marketed as "saving buyers" thousands on their purchase, this ignores the well-known issue that stamp duty is economically incident on the seller. 

What that means is that buyers have a willingness to pay for a home that includes the cost of stamp duty. If the market price of a home is $500,000 and stamp duty is 3%, then the purchaser must have been willing to pay $515,000 to buy the home. If stamp duty is removed, then this purchaser who was willing to pay $515,000 to buy the house can spend that $15,000 saved on stamp duty bidding up the market price. Instead of the government getting $15,000, the seller gets it. 

This is why a stamp duty holiday on housing purchases was enacted in the UK last year to "kickstart the stalled housing market".

So we know that reducing stamp duty alone will increase prices. But won't a land value tax decrease prices because it adds an additional cost for the buyer?

Yes.

And we are now at the crux of the issue. Neither tax will, in the short run, reduce the cost of housing. Homebuyers are paying the maximum they are willing. Whether that payment is directed to home sellers, or the government via stamp duty or land value taxes, has no effect on the underlying property market dynamics of rationing via prices and the costs buyers will need to pay.

In a presentation I made to Victorian Treasury a few years back, I explained this point with reference to the ACT's ten-year SD4LVT transition. I showed the below graph which makes the point that if the present value of the flow of future land value tax liabilities equals the stamp duty, the same willingness to pay for housing is just distributed differently under a land tax regime.

I also explained how to calculate the likely market price effect for a government replacing the same average stamp duty revenue stream with a land tax revenue stream.

The rule is

1. If turnover rate < capitalisation rate, price increases
2. If turnover rate > capitalisation rate, price decreases

Let's work through an example with some round numbers to demonstrate. The average housing sale is priced at $700,000 with a $37,000 stamp duty (5.3%). Without stamp duty, the price would be $737,000.

Assume that turnover is 100,000, or 4% of stock per year. This provides $3.7 billion in stamp duty revenue.

For a land value tax to raise $3.7 billion, it would need to raise $1,480 per dwelling, since in this example there are 2.5 million total dwellings. A land value tax rate in the dollar can be worked out in order to raise this amount based on the land value share of the average home. Say land value is 60% of the average home value, or $420,000, then the land value tax rate on the dollar is 0.35% per year.

Whether the market price rises or falls depends on whether the present value of a perpetual stream of $1,480 annual payments is worth less or more than $37,000 upfront stamp duty. 

If the capitalisation rate is 5% (the rate at which a flow of income is converted to a present value), then the present value of the land value tax is only $29,600. In this scenario, the market price would rise to $707,400 because a $37,000 stamp duty liability has been replaced with a $29,600 land value tax liability. Notice that the capitalisation rate here is above the housing turnover rate (5%>4%). 

If the capitalisation rate is 3%, then the present value of the $1,480 land value tax payment stream is $49,300. In this scenario, a $37,000 stamp duty liability is replaced with a $49,300 land value tax liability, and hence the house market price will fall by $12,300 to $687,700.

The question of whether SD4LVT is a good policy change depends not on the price effects—it merely redistributes who gets what payment and when—but on efficiency effects from making housing turnover cheaper. On that note, I refer you to my previous analysis showing that these claims are overblown and that the revenue volatility of stamp duty is a good thing because it stabilises the economy. 

In general, if you want to reallocate the economic rents that have accumulated to landowners, then a land value tax is a good way to go. But you do not need to remove another tax that achieves the same thing. Both taxes can work well together to divert economic rents from landowner to the public. 

Why I am anti-anti-zoning

I have fairly unique views on zoning, housing supply and prices. Here they are. Presentation slides covering similar issues can be downloaded from here. A video of the presentation is here.

Is zoning good?

The principle of zoning and land use regulation is very good and something I support. In practice, a lot of zoning and planning controls generally are designed poorly. I'm a big fan of simple rules.

For example, in Queensland, Australia, the developer charge (impact fee) on new housing was required to be set using a complex forecast of future population growth and infrastructure needs, with the fee per dwelling calculated in a big spreadsheet and published alongside various maps. I know it well. It was my job to help councils follow the required procedure and approve these modelled fees.

Then in 2011, the government announced a fixed cap on developer charges across the State. All councils quickly adopted this maximum charge. This system has been in place now for a decade, and it functions well enough. Its simplicity is a huge advantage, and across the planning system, there are similar gains from simplicity. (This surprise policy change also allowed me to demonstrate that these charges do not add to the price of housing)

I am a fan of the following process for managing land use.

• Periodic reviews (each decade) of local planning schemes. 
• Extensive community input into these schemes, perhaps with a citizen jury having the final say. 
• Reviews to require some minimum zoned capacity unless there are strong reasons not to (not every area needs to accommodate housing growth).
• Strict adherence to the scheme between reviews. 
• Few zones with simple controls (e.g. height, floor area ratio, setbacks, and use types).
• A simple, fast and cheap way for applications that meet the code. 
• A longer assessment for applications that do not meet the code, which allows for community input into the decision. 
• A mechanism for capturing value from changes of use (adopt the ACT betterment tax).

Across most of Australia, planning schemes generally follow this basic structure, though with limited value capture. Debates happen because buying land already zoned for what you want to build requires you to compete with other buyers and pay a higher price, reducing your potential profits. The big economic gains come from buying unzoned land then arguing about the planning scheme to get your site rezoned. Many developers choose to make planning applications with designs that do not meet zoning codes. By doing so, they choose the longer, slower, more expensive path. Doing so usually pays off for them if it results in more saleable building space on the site.

There is bad zoning and planning. I do not like systems where applications that comply with the zoning codes can be stopped. Yes, some bad designs will get through. But the whole point of creating the code is to "set and forget" the rules and not argue on a case-by-case basis.

I have no issues with detached house (single-family) zoning. I have no issues with car park requirements. Whether these are desirable should be debated during the review of the local planning scheme. In general, removing car park requirements should be accompanied by local transit investment and on-street parking management to address any issues. I'm not anti-car. I think cars are a good, flexible, way to move about in low-density areas. But they can be troublesome in high-density areas.

Generally, the transformation of detached housing into "missing middle" housing faces economic constraints. The rezoning jump should probably be to 4-5 storey buildings. Alternatively, allowing townhouse/ancillary dwellings that can fit in yard spaces will work to slowly increase density. New suburbs, even on the city fringe, should be allowed to have townhouse density housing, and the zoning decisions of these areas should fit with transport plans. I am not anti-suburbia. There is no problem with even large areas being zoned for low-density housing, especially if they are not planned to get transport upgrades.

Nothing is going to be perfect. While I think the planning profession overthinks things and tries to micro-manage more than it should, some coordination is necessary. Whichever way cities evolve, there are going to be arguments and conflicts. Planning can improve these non-planned outcomes and avoid conflicts if the system is kept simple and done well.

To summarise, I like good, simple zoning that includes plenty of zoned capacity and complements transport plans. I dislike complex micro-managed zoning systems. Unfortunately, most of the economic analysis of zoning never actually assesses planning rules, nor the zoned capacity in the system. Planning rules are usually just assumed to be a binding constraint on the total stock of housing. This type of detailed analysis is hard to do on a large scale because local rules vary. My recent study is one of the few that do.

I also do not think that shifting a city from a "bad" planning system to a "good" planning system has much of an effect on either the rate of new housing supply overall or on housing prices. 

I like more new housing supply. But the property market does not. To supply more than the market we need a non-market supplier. More on that later. 

Housing is an asset. It is priced like one.

Rent is the consumer price of housing. Only the rent is affected by the number of dwellings relative to the demand to occupy them. Rents rise with wages, not other consumer prices.

For the economists out there, think of consumers with Cobb-Douglas preferences in the utility function. This utility function generates fixed budget share results, where consumers spend the same share of their budget on each consumption alternative (on average across the distribution). This means that households will optimally spend roughly 20% of their income on housing, regardless of supply, and that the market gets to this equilibrium mostly through price adjustments, with higher-income households outbidding for relatively superior locations.

Here is what the private rental cost share of income is in Australia over a 20 year period. Yes, there are cycles, but in 2017-18 it was the same 20% income share that it was in 1997-98.

Most of the housing price adjustment over the past two decades has been yield compression—prices have risen while rents have been flat relative to household incomes. The housing asset, rather than earning a 5%+ yield when purchased at the old price, now earns a 2.5% yield. Prices double relative to incomes, but the consumer price of housing remains as expected, tracking wages.

There is no mechanism by which planning or zoning can affect housing asset pricing except via rents. This is why I do not take seriously the claims that Houston and Tokyo are good examples of how planning has constrained housing prices. Rents in those places reflect incomes, just like elsewhere, and both have had historically enormous price bubbles. In fact, Houston home prices increased the exact same 65% that Sydney prices did from 2010 to 2020. 

Density and supply are different concepts

A major confusion in the housing supply and zoning debate is that density, or dwellings per area of land, is assumed to be equivalent to supply, or new dwellings per period of time. This is wrong.

Developers choose the housing density at a site to maximise the site's residual land value. They also choose the rate at which they sell, developing to meet market demand in a "build to order" type of approach. This rate is known as the absorption rate. They also choose this rate of sales, and hence their rate of supply, to maximise the site's residual land value.

Many people seem to assume a contradictory idea that housing developers choose the density that maximises their land value but a rate of sales that minimises their land value.

I don't.

I've outlined how the optimal density and rate of sales work together in my paper entitled A Housing Supply Absorption Rate Equation. I take the logic of optimal density but include time in the model (yes, time is usually absent from models of new housing supply) to show how the two decisions relate. Here's a free version of the paper. Here's the published version. Notably, the main result is that the optimal rate of supply is independent of the density of dwellings. If it is optimal to make five sales per month when you are building 100 dwellings on a site, it is also optimal to sell five per month are building 1,000 dwellings on a site.

Rules on density do determine the location of different types of new dwellings. But the overall market rate of supply is determined by demand. There is no independent supply curve.

Undeveloped land is an asset

Undeveloped and zoned land is an asset that earns a return via capital growth. Owners of undeveloped land do not have to develop to earn an economic return. They manage ownership of these sites like any other capital asset that sits on their balance sheet.

The image below is from the annual report of one of Australia's largest housing developers. They own sites that can produce 56,000 apartments and 47,000 detached housing lots. But they aim only to produce 4,000-6,000 per year because that's what they expect the market will absorb.

I have a paper showing how undeveloped land is managed as an asset by housing developers rather than treated as inventory.

In fact, flexible zoning makes the value of undeveloped land grow faster before it is developed, incentivising development delay. This has been known since the 1980s when Sheridan Titman published a real options model of housing in a little journal called the American Economic Review. Titman's model relies on the idea that development timing choice will be delayed if density can be adjusted. If the optimal density increases with price, then sites with density limits will be developed first because the return to delay is lower. In density-constrained areas, the payoff to delay comes from higher prices, but in density-unconstrained areas, the payoff to delay includes both higher prices and higher density. A higher payoff to delay means a longer delay.

He writes

It is shown that the initiation of height restrictions, perhaps for the purpose of limiting growth in an area, may lead to an increase in building activity in the area because of the consequent decrease in uncertainty regarding the optimal height of the buildings, and thus has the immediate affect of increase in the number of building units in an area.

Vacant housing is also an asset—a more liquid but lower-return version of an occupied home. We can look to China for a good example of vacant housing stored as a balance sheet asset. Over 22% of dwellings there are vacant and housing supply has been astronomically high for decades, yet prices are still extremely high. Those high prices are due to the asset portfolio decisions of the Chinese people, which has compressed gross rental yields down to about 1.6%. Rents, as they do everywhere, reflect household incomes.

I recommend this video on Chinese housing. 

This is why I ask https://t.co/VY7C11iFbb pic.twitter.com/vHrBn38Sr8

— Cameron Murray (@DrCameronMurray) April 3, 2021

A problem with the standard economic approach to housing is that by disregarding the time dimension, the picture of the housing stock and its geographical distribution leaves no room for undeveloped land. All land is always developed to its highest and best use, and the only way the housing stock changes is via a complete demolition and rebuilding of the city at the new optimum. It is a model of the optimal density (economic frontier) not the actual stock or rate of supply of housing. 

The political economy

The political economy implied by anti-zoning advocates implies that well-connected property owners and developers are the weak ones, whereas the un-resourced Grandma who complains about development is the politically powerful one. Maybe that is true in some places. Not in Australia.

I did a study on rezoning and found that politically connected landowners are very likely to get their land rezoned and capture millions in windfall gains. Even our Prime Minister is a former property lobbyist. In these circumstances, are we really pretending that developers cannot build housing because of political barriers?

People also argue that when one site gets rezoned, the effect is mostly to increase the value of their land. But if a whole city got rezoned, then the value of all land decreases. Developers are playing a risky game by lobbying for city-wide rezoning when they actually want only their site rezoned. It is puzzling then that development lobby groups that call for large scale rezoning are so powerful. They should not exist in a world where large scale rezoning decreases land values, because no individual developer would be incentivised to support that collective lobbying effort.

If a whole city adopted a no-zoning rule, with no restriction on the density of housing or uses, what would happen? There would be quite a few sales as the asset market adjusted. Some landowners would sell up, others would buy in. Development in new areas where it was previously not allowed would happen. But after a short adjustment period, there would not be a noticeable change in the rate of new housing being built.

Auckland, New Zealand, a city of around 1.5 million people (about 600,000 dwellings) adopted a new plan that increased zoned capacity for dwellings by one million. There was no effect on the rate of supply. Can you spot when this happened in the below chart?

My view

Zoning is a useful tool for ensuring that different types of land uses happen in different areas. That's what it does. Just like lane markings on the road regulate where vehicles can go if they travel in different directions, zoning regulates where different types of stationary objects can go.

Zoning is not a good tool for slowing down, or speeding up, the rate of development. It is private landowners who make decisions about when and how quickly to develop.

If we want to build faster than the market absorption rate, we need a non-market housing supplier to do that. It is simply not optimal—either individually or jointly—for private landowners to sell new dwellings faster.

In fact, the recent low-interest-rate environment has made it optimal for private landowners to develop slower than otherwise. The logic is that developing housing is swapping a land asset for a cash asset. Since cash has such a low relative return, landowners are less in a hurry to do that. Higher interest rates would increase the equilibrium market absorption rate, but only after a short-term downwards price adjustment that will reduce the rate of new housing supply.

Lastly, I am suspicious of a political movement (e.g. YIMBYs) that argues that the market is not supplying enough housing but does not suggest any public intervention to build more housing. All the post-war success of boosting homeownership and housing supply was the result of heavy-handed government involvement in the housing market—rent controls, public finance of new construction, public land subdivisions, public housing, and more. All the problems we now face in a fully private housing market existed for centuries prior to the invention of zoning. 

As I like to say, the private property system is the original exclusionary zoning. All the problems I have seen blamed on zoning are merely problems of the private property system and will only be resolved by providing access to land and housing through non-market means. 

I would like to know what the people arguing for zoning reform as the secret to cheap housing and high homeownership think about how we historically achieved these outcomes?

1/4

— Cameron Murray (@DrCameronMurray) March 17, 2021

Making sense of property as a monopoly

Let us start by assuming that the property system is competitive and see how far that gets us in making sense of property pricing. After all, there are many different owners of different locations. That seems like what the textbooks describe.

The competitive market of economic theory has a few quirks. At the firm level, the demand curve is flat. Varying your own supply has no price effect. Yet, at the market level, varying supply does have a price effect. The problem of adding up a bunch of zero effects from firm quantity variation to a positive effect at the market level is an issue. It is resolved by assuming free entry at the market price; if a firm decreases its output while all others retain the same output, a new firm will enter the market and sell exactly that amount necessary to get back to the market equilibrium.

Neat, huh?

So let’s get back to it. We have a bunch of property-owning firms that can redevelop their space into housing. Each property owner sees a flat demand curve at the market price (because of our neat assumption) and has a cost curve that looks like...

Well, what does it look like exactly?

In strictly economic terms, the only input to the property right over a location is the existence of a property system. There are no input costs. To develop housing there are of course costs involved, such as fees, construction, and selling costs. But you can net these costs out of both demand and supply to look at the market for “empty” locations, or property rights to locations. After all, new housing supply is merely the subdivision of space—the subdivision of property rights into smaller pieces. The demand curve is still downward-sloping for that property rights market, but the supply curve must sit at zero.

Here we hit our first problem. If we assume perfect competition, we can only have land (location) prices of zero.

This “zero cost of location” view is what Ed Glaeser argues is the right way to think about housing.

…housing is expensive because of artificial limits on construction created by the regulation of new housing. It argues that there is plenty of land in high-cost areas, and in principle, new construction might be able to push the cost of houses down to physical construction costs.

In other words, land prices should be zero everywhere. Any deviation from this is due to regulatory intrusions in the market. Indeed, this implies that the very existence of a land market with trades at non-zero prices indicates it is not competitive. This is a bizarre conclusion in my view.

Maybe we will have more luck making sense of the property system by assuming it is a monopoly despite the many different owners. This has some intuitive appeal. First, you can’t choose to have your locations supplied by a competing property titles system. You can’t run more than one property system in parallel. Can you imagine the conflicts over who owns what space with multiple property systems? Second, the ownership of monopolies is often carved up (subdivided) and owned by many different people who each own constituent parts. Though we usually call these company shares or stocks. 

So what then of the economic theory of monopoly?

As a starting point, a monopoly model avoids the assumption of free entry as the reason individual firms cannot observe their own-supply effect on price. This matches the reality of the private property system as it does not allow free entry. You can only compete in the property market by first buying property from the property market. This seems sensible.

The property market for any use, like housing, therefore looks this at a point in time.

Since costs are all sunk, we can simplify a bit by ignoring the stock of existing housing and look forward in time only. Think about redrawing the axes with the origin at the equilibrium current price and stock.

A change in the housing stock is the supply of new housing. The question of interest is how the stock of housing evolves over time in a property monopoly to determine an equilibrium rate of new supply?

In this model, if demand stays fixed and existing housing does not depreciate, then no new housing is built.

As demand shifts, new supply is added at a rate that maximises the revenue gain from selling those new properties rather than keeping them undeveloped in your balance sheet. It is the same monopoly maximisation principle applied at the margin. In other words, you sell new housing lots at a rate that maximises the present value of that flow of sales.

This logic is shown below. From the t=0 equilibrium, demand shift upwards. The green line shows the marginal revenue from the new supply (i.e. the change in the housing stock, Q, over that period). The new equilibrium is where the revenue from that flow of new property put to housing uses is maximised (i.e. where marginal revenue from that flow is zero).

This equates to a rate of supply that I explain in my absorption rate theory of housing supply (assuming a zero interest rate to simplify the inter-temporal trade-off).

The steeper (more price sensitive) the demand, the less supply responds to the same vertical demand shift. That’s because each property owner is sensitive to their bigger own-supply effect on the market price. Thin markets mean less supply.

Under this monopoly logic, the supply curve for property is not an independently-determined cost curve, but a derived curve based on the slope of the market demand curve.

But how do you get to the monopoly outcome? Is a conspiracy needed?

Not at all. All that is needed is trial and error. Remember, unlike the competitive market assumption of free entry, in the monopoly model, changes in firm output affect market output. If firms start near the competitive rate of supply per period in the face of a sloping (and rising) market demand curve, they will quickly learn to get to the monopoly rate of supply independently.

All that is needed is a learning rule of “win-stay, lose-shift”. This rule says that if the increasing the rate of supply last period increased your marginal revenue, then increase next period, otherwise decrease the rate of supply. If decreasing the rate of supply last period increased your marginal revenue, continue to decrease, otherwise increase.

When I simulate this learning rule with three firms, they quickly converge to the monopoly rate of supply, and hence the monopoly price. This convergence will occur even with 1,000 firms in the market. Property markets have been around for a while now. It seems likely that the current set of property owners has learnt this maximising equilibrium.

It seems logical to conclude that the monopoly model of property makes more sense than a competitive model. Indeed, monopoly was the traditional economic way of understanding property markets. In terms of understanding housing markets and effective policies to reduce prices, I think the following points are key.

1. That the property system is a monopoly shows that the rate of new housing development is mostly a product of demand. Property owners don’t want to build faster. This is because supply is not independent of demand. Supply is merely a reflection of demand.
2. Rezoning and changing planning rules might change where development happens (as it should—they are location regulations after all) but probably won’t change the rate of new supply. The rate is the one the market wants. Developers don’t want to flood the housing market.
3. Making housing cheaper should be understood the same way we understand other monopolies. We regulate prices. We create public options. Pretending that we can somehow capitalise on competitive market forces that don’t exist won’t change things.

Luxury beliefs of YIMBYs and free markets

Yes In My Back Yard (YIMBY) is the belief that more housing density in your neighbourhood is good. But I argue that almost nobody who makes these arguments truly believes them. Mostly, it’s a case of Yes In Your Back Yard (YIYBY). I say that because even property developers who make a living building high-rise apartments hate it when these same types of apartments are built in their street. What about the politicians who impose massive density increases in the suburbs? They hate it when that policy arrives in their street. Sometimes YIMBYs will change their story when density comes to their neighbourhood to one about "missing middle density" or "urban design". Yet if their belief in the economics of more supply was genuine, it should trump these minor concerns.

Former pro-development Brisbane Mayor turns NIMBY when not-for-profit aged care comes to his neighbourhood. https://t.co/pHz65YD4dy

— Cameron Murray (@DrCameronMurray) March 12, 2021

Such hypocrisy also happens with immigration policy. Those who support turbo-charged immigration policies do so because they personally are insulated from the effects. But when those effects come to their own leafy town, they turn their beliefs 180 degrees to be against it. All those apparent benefits used to justify the belief exist only when others bear the cost.

These are classic examples of what Rob Henderson has called luxury beliefs. Things you don’t really believe, but use to signal group loyalty and status. Luxury beliefs are “ideas and opinions that confer status on the rich at very little cost, while the main price is paid by those less fortunate.” 

In economics, I think the efficiency of markets is a classic luxury belief. Even the most pro-free-market economist, like Milton Friedman, are out there speaking for status, rather than reflecting true beliefs. Friedman might even have himself fooled most of the time.

For example, when speaking to a group of Chicago economics colleagues, he “was bemoaning the fact that Chicago price theory was dying out.” One colleague responded “Milton, I thought you believed in markets. It sounds to me like price theory is losing.” At least according to Steven Levitt who said that “even Milton Friedman, in the end, didn’t really believe in markets when markets moved against him” (from 52.00 here).

In fact, markets are the last resort for resource allocation. As a society we really hate markets. We avoid them as much as possible by creating firms to bring market transactions within a hierarchy, and when we have markets, we regulate them heavy-handedly to ensure they function in a very un-market-like way. Only those expressing luxury beliefs who are insulated from the costs pretend otherwise.

I would even go so far as to say COVID policy has been fuelled by these luxury beliefs. Pro-lockdown politicians repeatedly violated their apparently strong beliefs when it came to their personal lives. Perhaps they were expressing status and loyalty with their lockdown comments, not their genuinely held beliefs.

The fact that people say things that contradict their actions is one reason why many economists are sceptical of survey responses. You can ask people “is it a good idea to build more dense housing in this neighbourhood” but then those who agree will end up spending thousands on lawyers defending the status quo situation in court. Their response was not genuine. It was a luxury belief. Better to ask survey questions that are more personal, like "it is a good idea to build high-density housing next door to your house?"

What can be done to weed out luxury beliefs?

One thing is to align decision-makers with the costs of policy change. Put them personally on the receiving end of the costs of the policy and you will see their true beliefs emerge. High immigration and housing density in the suburbs where politicians live needs to be a prerequisite for the same policy elsewhere. Alternatively, we could randomise the decision-makers through sortition—a type of random jury system to become a politician—ensuring that those who bear the costs of a policy are part of the political process. 

Evil rent control revisited

1/8 One year ago, the left-leaning government of Berlin introduced unprecedented rent controls.

Lots of economists warned against it.

So what do the numbers say now?

A thread. https://t.co/wIDm9awhre#Mietendeckel @bopinion

— Andreas Kluth (@andreaskluth) March 2, 2021

A new report is out looking at the Berlin experience with rent control (original German report here). The basic idea is that dwellings built prior to 2014 now have regulated rents and caps on rental increases. All very standard price-cap regulation. New dwellings are not regulated. 

What I find funny are the contortions that economists make to explain good things as evil things because of their cultural taboo about rent control. With the help of another recent paper on rent control, I want to straighten out some of these contorted arguments.

Rents in the unregulated market don’t fall. This is evil. 

High rents are bad. This is the foundation of the claim that rent control is bad. Yet the enormous benefit of rent growth at a rate 60% less for roughly half of all households in Berlin is ignored and downplayed.

There aren’t many other policy interventions that get this size outcome. Even doubling home building for a decade is at most going to see a 10% relative rent reduction in a decade's time, which in terms of growth rates would be a tiny effect. This rent control case study appears to be one of the most successful affordable housing policies I've ever seen.

Even if you assume that market rental increases of the past year are related to rent-control policy (rather than being a continuation of previous market trends, which my eyeballs tell me is the case, especially in the original report chart with error bars marked) the welfare calculation is hugely in favour of rent control.

The basic welfare calculation is:

With a 0.5 share of households renting pre-2014 buildings with rent rising 2% per year instead of, say 5%, saves each household about half of the current year’s rent over 5 yrs. That’s 0.25 “city rent years” of benefit to residents.

With a 0.05 share of households relocating each year paying at worse 0.08 more for their rent, we get 0.027 of “city rent years” of cost to residents over 5 yrs.

On pure "high rent is bad" benefit-cost terms the Berlin experience got a 10x economic return! This even assumes that all private market rental growth is due to rent control, which is implausible (more on new supply later).

You can’t just pretend that these orders of magnitude are similar.

Additionally, the macro-economic effects of additional non-housing renter spending in the local economy must be quite large. This is a diversion of half the rents in the city from landlords to tenants, with the downwards income distribution likely having large spending effects.  

Landlords sell to homeowners. This is evil.

“Landlords treated by rent control reduce rental housing supplies by 15 percent by selling to owner-occupants and redeveloping buildings.”



I find this statement hilarious. 

Landlords sold to owner-occupants. This is a good thing, not an evil thing. 

Many policies are aimed at getting more homeownership and the best way to do that is for landlords to sell to owners (even the tenant perhaps). 

If they do this, there is no effect on the supply of dwellings. This is because the former renter who buys the home is now also no longer a renter. It’s a minus one from the supply AND a minus one from the demand. 

Redeveloping buildings increases new housing supply. This is evil.

The whole economic schtick about rent-control is that it decreases new housing supply. But both the Berlin report and the Diamond et. al. paper show this is not the case. 


Diamond shows that rent-controlled housing is roughly 5% more likely to get additions and alterations, and 7% more likely to be redeveloped. But I thought rent control meant that landlords won’t invest in their buildings anymore? The charts below show these effects over time.

In fact, there is a whole movement of economists and urbanists who are itching for ways to get more housing supply because they believe it helps reduce the cost of renting in prime areas. Maybe try rent control?

Turning now to the supply of unregulated rental housing advertised in the market, it looks from the Berlin study that new apartment listings are outpacing peer cities. But this is labelled evil. They are "only slightly outpacing" other cities. Terrible!

Rent control stops poor people from being displaced. This is evil.

The final strange contortion happens in the Diamond paper when concerns over low supply (now invalidated) turn into concerns over gentrification and the rapid new supply of luxury market-priced housing. That rent-control works to stop the displacement of poor people during this process is then labelled as evil by saying that it widens income inequality. 

…it appears rent control has actually contributed to the gentrification of San Francisco, the exact opposite of the policy’s intended goal. Indeed, by simultaneously bringing in higher income residents and preventing displacement of minorities, rent control has contributed to widening income inequality of the city. 

The problem with rent control is that it works. This is evil.

The real issue is not that rent-control is an economic failure. The zeal with which the theoretical economic case against it is made, regardless of the evidence, is necessary because rent-control works. Without an economic story for why rents should not be regulated, all those landlords would constantly face the political risk of losing billions in rents. 



One argument that I’ve heard is that economists have tricked themselves with their own supply and demand theories. But I think that’s wrong. I argue that the theory is used to back-fill a convenient political story. I say that because supply and demand logic can just as easily be used to show why rent-control should be a huge success with no changes to investment incentives around new housing. 

Let me show you. In the first supply and demand diagram below I include total housing supply and demand, with a kinked supply curve to represent the existing stock of dwellings that don’t get demolished because prices might fall. This the way Ed Glaeser describes housing supply curves.  

All we then do is split the demand curve into existing and new demand. We subtract out the tenants of existing houses from new demand, and their existing homes from new supply, and are left with the following, where the dashed blue line is the demand curve for existing housing and the solid blue line is the demand curve for new housing. It’s the same equilibrium. We merely cut off all demand and supply to the left of the market equilibrium and put it in the regulated system.

The economics of rent control is evil.

The taboo topic of rent control is so screwed up that the evidence it works must be contorted into something evil.

And that's all folks. 

Merit doesn't cause income

In a podcast interview with the terrific Joseph Walker, geneticist Robert Plomin said something that stuck in my mind—"in a meritocracy there will still be large deviations of income due to genetic variation in abilities."

I also stumbled onto an economics blog about capital taxation that relied on the idea that high capital incomes come from previous savers who therefore deserve that higher income (I can't seem to find it again, but I'm sure you've heard the idea before). To bring Plomin's comment into the mix, we could say that if genetics determines time-preferences or risk-taking attitudes, then capital incomes are also determined genetically. Taxing those capital incomes is therefore taxing the "merit" of saving.

Like much of the economic and political debate about poverty and inequality, these ideas rely on a flawed understanding of how income is distributed in human societies. That flaw is to pretend that there is something called "merit" or "value" (or "productivity" for that matter) that is independent of the ability to be rewarded via payments.

Think about it like this. We often look at the characteristics of people who earn more money—their education, their patience, their leadership skills, their attractiveness—and then label those attributes that correlate with incomes as "merit", or "value", or "productivity". This approach merely defines "merit" based on current income. If "merit" causes incomes, there must be a way to define and measure it independently before looking at any correlations.

Imagine LeBron James was born in 1784. His genetic gifts are identical—his athleticism, work ethic, and his overall intelligence on every dimension. His genetic "value" or "merit" assessed without reference to his income is identical to what it is now.

But society at that time would be structured in a way that his genetic gifts would make him a high-value slave. His personal reward would be the pleasure of working hard manual agricultural tasks for longer, and the slave owner would get the income for having the "merit", "value", and "skills" of cultivating high productivity slaves. (Do you see the parallel with capital owners generally? Their "skill" and "value" is choosing investments. In practice, that means choosing what tasks other people do when they work for their income.)

Here's another example. In-breeding of royal families historically led to genetic variation that made many of them useless in the productive activities of the day. Despite this, their incomes and power remained nearly identical to their genetically superior siblings. If our social structures mean that "merit" determines income, and "merit" is heavily genetically determined, this should not happen.

In fact, if you follow the logic that genetic variation causes merit variation, which causes income variation, then genetically diverse places should naturally have more inequality. That "homogenous Nordics" argument rears its head again. Compression of genetic variability or some other attribute we have labelled "merit", like education, is therefore seen as the best way to compress the income distribution.

But in reality, the Nordics compress their income distribution through welfare state policies. Their market income distribution is much the same as peer nations. They merely choose to create a social structure that keeps the income distribution more compressed regardless of any underlying variation in individual attributes, genetic or otherwise.

We also know that income and wealth inequality has grown tremendously in the past four decades. This has nothing to do with changing genetic variation. It has nothing to do with a changing distribution of an independently-assessable concept of "merit". School teachers make lower relative salaries today because we chose not to raise their salaries faster. Some countries didn't do this, and their school teachers make more money. 

Some of you might be thinking that places with high-income school teachers have better teachers. But you are again failing at basic logic by reversing out "merit" from income. 

We know that increasing incomes actually causes changes in personal attributes that are usually labelled as "merit" and assumed to be determined independently from income. Basic income experiments show that higher incomes cause people to become more confident, trusting, healthy, less depressed and more able to concentrate, for example.

It seems logical to me that if there is no such thing as an independent concept of "merit" or "value", then we should simply choose a more compressed income distribution in society rather than the unequal one we have chosen. That more compressed distribution will help avoid the pitfalls of high inequality by ensuring that regardless of variation in genetics or "merit", people can live comfortable lives and take opportunities that low incomes often prevent them from doing. 

In practice, this comes about by setting an income range via the tax and transfer system. Very high marginal tax rates on high incomes, coupled with generous unconditional payments for low-income households, will compress incomes.

Perhaps the minimum income could be roughly $20,000 per person (with variation depending on household size and type), with a maximum of $2,000,000 per person. That leaves a 100x income range within which market-based incentives can operate. I see no reason why the range of variation should be any higher than that. Can individual merit, however defined, really explain a 1000x variation in income? Will some people not go to work because they earn $8,000 per day instead of $20,000 per day, even if there are no $20,000/day options because of the tax system? I doubt it. People won't even see a change in their rank within the distribution.

Income compression should be the focus of policies for dealing with poverty and inequality. Ideas that rely on "merit", "value", or "productivity" are going to be failures as they implicitly (or explicitly) define merit by income. 

The Henry George logic of wages as an economic rent

George was concerned that property owners gained all the economic rents. But his logic relies on a quite arbitrary and limited concept of property rights.

We know, for example, that human slaves could historically be the exclusive property of another person.

By George’s logic, the economic gains to human-type property owners would be an economic rent, just as the gains to location-type (or land-type) property owners are an economic rent.

From the viewpoint of property and economic rents, freeing slaves transferred a property right, in the form of the right to choose labour tasks and receive payment for them, from the slave owner to the slave.

Thus, the wages of freed slaves are the economic rent from the property right they now own. The owner of human-type property still gets the rent. It is just that each person became an own-slave-owner.

Hence, when modern Georgists say things like “all taxes come out of rents” they are just saying “all taxes come from property rights owners who can capture the economic surplus.” Or simply, “all taxes come from someone.” 

But because the understanding of property rights is limited mainly to location-type resources, not human and other resources, they miss a big part of the economic picture. While it is certainly the case that owners of location-type property need not labour for their share of economic gain, changes in the distribution of the ownership of location-type property can spread those gains more widely, just as a change in the distribution of slave-type property owners spread the gains from human labour. 

George's idea of a tax on the value of location-type property is a good way to socialise the gains from concentrated land ownership. In the modern era of high top labour incomes, taxes on the high value of these human-type property rights might also be a good way to socialise some of these economic rents.

The balance of wages, profits and rents, are not a product of physical aspects of production, but rather than property rights distributions. George somewhat indirectly makes this point when he writes:

Where land is free and labour is unassisted by capital, the whole produce will go to labour as wages.

Where land is free and labour is assisted by capital, wages will consist of the whole produce, less that part necessary to induce the storing up of labour as capital.

Where land is subject to ownership and rent arises, wages will be fixed by what labour could secure from the highest natural opportunities open to it without the payment of rent.

Where property rights to land are limited (i.e. land is free), wages and capital owners get all the surplus. As one would expect, only those with property rights can get a share of the surplus. If then property rights to capital were removed, or limited, then all the gains would go to labour. We can get any economic distribution we like with different sets of rights (aka rules). 

The relationship between George's insights on rents and the concept of property rights need more careful consideration in the modern world where the evolution of property rights tends to be invisible to many economists and policymakers. 

Downs-Thomson housing paradox

Anthony Downs and various others made the observation that “the equilibrium speed of car traffic on a road network is determined by the average door-to-door speed of equivalent journeys taken by public transport.”

Though known as the Downs-Thomson (DT) paradox, it is not a paradox at all. It is merely an observation that people adapt their behaviours to road network changes that affect the cost of driving.

The basic idea is that there are three margins from which substitution towards peak hour use of new freeway capacity occurs.

1. many drivers who formerly used alternative routes during peak hours switch to the improved expressway (spatial convergence), and
2. many drivers who formerly travelled just before or after the peak hours start travelling during those hours (time convergence), and
3. some commuters who used to take public transportation during peak hours now switch to driving, since it has become faster (modal convergence).

This is a simple application of the logic of substitution in microeconomic consumer theory. If an alternative has a lower economic cost for the same value gained, people will substitute towards the lower-cost alternative. Consumer choices are what enables markets to select for the more efficient producers and mix of goods and services over time as they respond to relative prices. 

In the DT example, the three alternatives and the peak hour freeway use start at the initial equilibrium. Prior to the new freeway, all the transport alternatives are taken up to the degree that the total cost (time/money/convenience) is equalised across them.

The cost of travelling on an alternative route, or at an alternative time, or using an alternative mode, is roughly the same as the cost of peak hour freeway transport. It must be because if it wasn’t there would be a substitution towards the lower-cost option.


If you reduce the cost of one of these alternatives what happens? That option begins to look relatively attractive, and people substitute from the other alternatives until there are no remaining gains from substitution.

Afterwards, the new equilibrium looks like this. I’ve put a dashed line at the old equilibrium to show that there can be overall gains from expanding transport capacity, even though they are not observed in the option that received the investment.[1]

How much below the previous equilibrium the cost profile of the transport alternatives remains depends on how much these costs can adjust downwards and the preference for substituting towards non-transport uses of time and money.

For example, if alternative transport modes have a fixed cost (e.g. taking the train in peak hour has a fixed price and time cost that does not change based on usage), then the new cost equilibrium will occur at the same level as previously. That fixed cost of the alternative mode will anchor the equilibrium time and cost level at which these options equilibrate.

This is what is meant by the observation that “the equilibrium speed of car traffic on a road network is determined by the average door-to-door speed of equivalent journeys taken by public transport.”[2]

Similar margins of substitution happen in housing. Housing is another spatial allocation problem where the cost involves a trade-off between price and time costs amongst alternatives.

Location alternatives are one such margin. Total cost equilibrates, via rent adjustment, between comparable dwellings at alternative locations based on relative transport/accessibility costs.

When the transport/accessibility cost of a location falls, the rent in that area rises so that there are no gains from substitution between locations.



Another margin of substitution in housing is between modes, like renting and buying. When the cost of some buying goes up, it can drag up the cost of renting as market participants close the gap between buying and renting. Alternatively, if the cost of renting falls, it tends to pull down the cost of buying.

What if there is another alternative housing mode? Say, a social or public housing option that has a fixed price. In this case, the point at which there is no substitution between them is where they all have the same economic cost.

If the cost of public housing falls, this puts pressure on the housing equilibrium. Homebuyers and renters begin substituting to this cheaper alternative, just like the case of the new freeway in transport equilibrium.

This substitution process continues until there are no gains from substituting between housing modes. Public (non-market) housing alternatives can create an anchor for prices, just as public transport alternatives anchor congestion levels.

One could argue that “the equilibrium price of housing in a private market is determined by the average price of equivalent public housing.”

Our policy choice not to extensively provide cheap housing alternatives has allowed housing prices to be anchored by the maximum willingness to pay in the market rather than the cost of public housing.

The DT paradox is why places with low traffic congestion are usually those that have good alternatives. It is also why places with cheaper private housing markets have cheaper and more widely available non-market alternatives.

[1] This is actually how monetary policy works. The overnight cash rate is manipulated, and then all the alternative interest rates shift because there is a substitution between the overnight rate and long-dated assets.

[2] The D-T paradox has implications for pricing choices on transport networks. For example, peak-hour congestion charging will raise the cost fo driving, creating substitution to alternatives, including the alternative of not travelling. When a new rail line is built to relieve pressure on roads, tickets must be priced low enough to attract people towards that alternative and away from road travel. If new rail lines are privately owned and earn a return from ticket sales only, they have an incentive to maximise their own revenue, but this may involve a ticket price that is too high to maximise overall transport gains by attracting more substitution away from road travel. New toll roads also are likely to over-price compared to the optimum for the network as a whole that would see more substitution to these new roads.