Construction costs DO NOT increase housing asset prices, they SHAPE the density and quality of the pool of feasible homes (Part II)

Part II of a series on why we don't know if housing construction productivity is rising or falling, and even if productivity was falling, why that won't increase housing asset prices

As always, please like, share, comment, and subscribe. Thanks for your support. You can find Fresh Economic Thinking on YouTube, Spotify, and Apple Podcasts.

Interested in learning more? Fresh Economic Thinking runs in-person and online workshops to help your organisation dig into the economic issues you face and learn powerful insights.

---

In February 2025, Australia’s Productivity Commission (PC) released a report about a decline in measured labour productivity in the housing construction sector and made many strong policy recommendations based on its analysis.

Unfortunately, that analysis was lightweight and ignored important economics. In Part I, I asked this question: “What are we measuring and what would we expect that measure to show based on economic logic?”

Read that here.

We DON'T KNOW if housing construction productivity is RISING or FALLING (Part I)

A key part of the analysis in Part I showed that if labour productivity gains are made in what we categorise as the housing construction sector, these more productive enterprises, such as prefabrication facilities, are classified as manufacturers and no longer fit in the housing construction sector.

Now, I dive further into the question: “Why don’t higher construction costs increase the asset price of homes?”

The full article is for paid subscribers. Please consider upgrading today.

PART II - Productivity is not price

The story so far is that there are good economic reasons not to interpret a decline or increase in the measured sector-specific labour productivity the way Australia’s Productivity Commission (PC) and many others have done.

The observed labour productivity trend in housing construction is a global one, suggesting that underlying economics, rather than country- or city-specific regulatory systems, are the primary cause of the measured decline.

A big part of why such measurements fail to capture reality is that when there are productivity gains in construction due to mass production or prefabrication, those activities get reclassified as manufacturing, leaving the construction industry looking like a productivity laggard. Economist William Baumol explained why we should expect industrial productivity inter-dependencies like this more than half a century ago.

That is not to say that the construction sector does not have areas for economic improvement and productivity-enhancing investment. I certainly support experimentation with prefabrication, relaxing or removing outdated regulations, and more. Let’s push and see what we can do. Incremental efficiency gains are good wherever they can be found.

But remember, the regulations we have exist not by chance but because of the historical problems in this high-stakes credence good¹ construction sector. Sure, they may be outdated, and the benefits might not be worth the cost. We should work on that.

The strong claims made in the PC report go far beyond whether labour productivity was rising or falling in housing construction. They also imply that a) high productivity leads to lower housing construction costs, and that b) lower housing construction costs lead to lower housing asset prices, particularly via faster rates of new housing production per period.

I explained why the first claim was wrong in Part I.

But why isn’t the second claim true? Surely, lower construction costs mean lower home prices and rents? To ask the question another way, if construction costs fell by half due to some fancy new tech, what would the effect be on housing rents and asset prices?

Here, we must start by asking where the price of housing assets and land comes from. Then, we must look at how property owners choose both the optimal density and quality of a development project, as well as how the optimal rate of new housing production emerges across the market of interacting property owners and buyers.

The main insight here is that differences in construction costs don’t determine market prices, nor the rate of production. Instead, they determine a project’s optimal housing density and quality at market prices. We don’t take productivity gains directly in the form of lower housing prices. Mostly, we capitalise on productivity gains by changing the density and quality of homes built, and in the form of overall income gains in all economic sectors which allow us to pay higher rents and prices in desirable locations. This was William Baumol's famous insight noted in Part I.

Where do housing asset prices come from?

It seems a simple question. But few ask it.

Many confuse the cost of building a home with its market price. But since the same home with the same construction cost built in two different locations will have a different market price (and a different rental price), this cannot be the case.

It is important to separate the two prices of residential property—the rent, which is the price of the housing service, and the asset price. I won’t discuss rental prices here, but I will refer you to this article that explains how these prices arise.

Here’s how I have previously explained the economic forces that determine housing asset prices.

Most people would say [housing asset prices come from] supply and demand. That’s a start. But it doesn’t tell us much more than that sellers and buyers agreed to prices.

I mean, who else sets the price?

Supply and demand is merely a description of market price adjustment. It doesn’t tell us why buyers were willing to pay a certain price, nor why sellers want to sell and are willing to accept that price. It doesn’t tell us anything about the broader equilibrium trading forces in the market.

The trick to understanding property asset market prices (yes, housing is an asset) is that every dollar spent on a home can be spent on any other asset in the economy. Homes can’t be priced for long at a level whereby the return on a dollar invested in housing is far out of whack with the rates of return on a dollar invested elsewhere.

The rate of return from rents and capital gains on housing must track closely the prevailing rates of return in other asset markets over the long term.

This is why monetary policy, whereby the interest rate on bank deposits is lowered or increased, has a substantial effect on home buying and prices.

Lower returns on cash assets make the returns on housing at the prevailing price seem more attractive. To use the checkout analogy, money leaves the cash queue and goes to the housing queue. Vice-versa when interest rates rise (though there are longer lags and more price stickiness when it comes to declining asset prices).

An easy way to think about this equilibrium relationship with other assets is that homebuyers can either rent a home from a landlord or rent money to become their own landlord.

To reiterate, for a given level of risk, total returns must converge between assets. If they don’t, there is an arbitrage opportunity to sell one asset providing a low overall rate of return per dollar invested, and buy the higher return asset.

Total returns include net incomes, like rental income or dividends, but also returns in the form of asset price increases, known as capital gains.

The diagram below illustrates the case where returns on housing assets are lower than the returns available elsewhere, given the same risk. In this situation, people will sell housing and buy other assets, causing the asset price of housing to fall, other assets to rise, and these two rates of return per dollar invested to converge.

It is also the case that there are low-risk assets and high-risk assets. Low-risk assets will generate lower returns relative to higher-risk assets—a well-established risk-return relationship across all financial markets.

That relationship is also an equilibrium, or frontier—you can get close to it, but only exceed it briefly before trades restore the equilibrium due to arbitrage opportunities. The below chart demonstrates.

What does this chart mean, and how does housing fit on it?