Anecdotally, the story often goes like this:

A land owner leases his commercial retail space at an expensive price. The place stays vacant for months or years at a time. Eventually someone decides to lease the place. The rents are very expensive and eventually the leaser goes out of business.

There are many other families like the Wakils, who are content leaving their property vacant. This example is largely relating to Australia but it does happen elsewhere.

Why would they do that? Surely any rent earned would be better than no rent.

My thoughts are:

  • There could be some sort of tax concession in place. If a landowner tries to lease for \$5k a month, the tax write-off could make the owner better off than if they actually leased it for \$4k a month. The below suggests this happens to at least some degree.

    Leichhardt mayor is calling for a range of legislative changes to allow rates to be increased and tax concessions reduced for landlords who leave their shop fronts empty unnecessarily... there is currently no allowance to reduce rates for those property owners who keep their properties tenanted or to increase rates for landlords who are using their properties as a tax write-off.

  • The returns to land-price speculation make leasing a secondary priority.

  • Structural change in how valuable retail space is. With the increase in grocery deliveries and other forms of online shopping, maybe people don't have as much need to shop. This, coupled with possible stickiness in rental prices will result in a significant drop in quantity demanded.

  • Some people just don't care. (I find this particularly unsatisfying considering the magnitude of the values involved)

To me, it is clear there is at least some form of regulatory failure present here. If these places were rezoned into residential, they would get filled very quickly as this phenomenon seems primarily restricted to commercial real-estate. The council (and probably the community) would prefer to have this area remain commercial but residential is surely better than vacant.

  • 6
    $\begingroup$ In many jurisdictions there are restrictions on when and how a landlord can increase rents. If you let at a price far below market levels today then you may be locked into a stream of low rents for a long time. $\endgroup$
    – Ubiquitous
    Commented Aug 3, 2015 at 9:11
  • $\begingroup$ wrt the bounty, looking for an answer which explains this phenomenon clearly. A perfect answer would include the intuition and have solid references. $\endgroup$
    – Jamzy
    Commented Aug 5, 2015 at 0:38
  • 2
    $\begingroup$ Tenants may fail to pay. They may also cause damage to the property. In some locations, it may be difficult to evacuate such tenants, either because of strong legal protection for tenants, or conversely because of weak local police. In such cases, letting your house for rent is very risky. $\endgroup$ Commented Aug 5, 2015 at 13:52
  • $\begingroup$ I am getting increasingly sure that this won't get answered. I spoke to a top property economist the other day and his response was 'vacancy is a huge to want to avoid in property... I guess some people are just jerks' $\endgroup$
    – Jamzy
    Commented Aug 10, 2015 at 23:06

3 Answers 3


Before applying irrationality to explain the phenomenon, consider the following simplistic model of a rational owner: assume there is the high rent $R^h$ and the low rent $R^{\ell}$. The owner faces the prospect of earning either

$$V(h) = \sum_{t=1}^{T_1} \beta^{t-1}R^h + \sum_{t=T_1+1}^{T_2} \beta^{t-1}\cdot 0+ \sum_{t=T_2+1}^{T_3} \beta^{t-1}R^h +...$$


$$V(\ell) = \sum_{t=1}^{\infty} \beta^{t-1}R^{\ell} = \frac{1}{1-\beta}R^{\ell}$$

Favor the "low-rent" scenario by ignoring all terms in $V(h)$ except the first sum (this reflects that the property may go unleased for "many years", so remote-future income gets fully discounted). Based on these considerations alone, if we observe the "high rent" scenario, it must be the case that

$$V(h) > V(\ell) \implies \frac{1-\beta^{T_1}}{1-\beta}R^{h} > \frac{1}{1-\beta}R^{\ell} \implies R^{h} > \frac{1}{1-\beta^{T_1}} R^{\ell}$$

which can be re-arranged to give

$$\beta < \left(\frac {R^h-R^{\ell}}{R^h}\right)^{1/T_1}$$

The right hand side can be estimated by real world data: obtain average magnitudes for the high and the low rent, as well as the average time a business under the higher rent pays it prior to fold (the $T_1$ parameter). These will lead to an estimation of the upper bound for $\beta$ in order for the above inequality to hold, and justify the "high rent" scenario observed.
Then one can examine this upper bound and contemplate whether it appears "reasonable".
As a numerical example, assume $R^h = 2R^{\ell}$ and $T_1=2$ (say years). Then it must be the case that

$$\beta < (1/2)^{1/2} \approx 0.71$$

"Too low" (i.e. too high discounting) one would tend to say. But don't look at benchmark values of $\beta$ that are used in macro-models. Try to think what the discount factor attempts to capture here. And in any case, a person with very high discounting, is not "irrational" -this is how he experiences the world.

And then, there are other considerations: accepting low rents tends to drive the asset values down. Also, It may be the case that this is incremental income for land owners, above some comfortable level from other sources, meaning declining utility of wealth may kick in. Further, constant occupation physically wears the asset down faster. Etc.

My point is, "people are just jerks" (i.e. arguing for irrationality) can explain everything - and so it really explains nothing.


Two economic models and one returns calculation that can help provide some intuition for this issue and maybe even get to a quantitative evaluation of the problem if anybody finds the right data:

Option exercising: The theory of options suggests that you don't exercise a call option whenever the underlying is above the strike price because.... you are foregoing the chance that the underlying goes even further above the strike price! So a vacant property is a property whose owner is waiting for an even better occupant than the ones it has found. Once you rent it to somebody you are locked in (many times by law) for potentially many years: you can't show it, you can't sell it, you can't remodel it, you can't raise the price as you wish, etc. So in a volatile environment, or a situation of a changing neighborhood, a landlord will be willing to wait a long time for the right tenant.

Search models: In many economic scenarios there's a market where every buyer and seller are slightly different, and it is costly for them to search for the best match. It takes them time and the likely success rate depends in part on the ratio of buyers to sellers. In a scenario with many sellers, it might take a long time for an individual seller (owner in this case) to find the right tenant. He might even be discouraged from searching for a tenant and prefer to wait it out.

Returns calculation: Lastly, you have to realize that real estate investors are getting returns from two sources: R= (P(t+1)+Rent)/P(t). In other words, there are two sources of return to investors, the change in price and the rents. The fact that a site is vacant does not mean the owner is loosing money, he might be getting a fast-enough appreciation rate. Moreover, its possible to think of circumstances when P(t+1) rises faster if the property is empty. For example, when the owner foresees a big redevelopment firm coming in and wanting to buy his property. It better be free at that point or he would have to delay the sale for months of years.

  • $\begingroup$ Welcome and (+1) for the suggested interpretative frameworks. I see that you are a moonlighter, so good luck with your night job. $\endgroup$ Commented Mar 11, 2016 at 22:34
  • $\begingroup$ I know it's been a long time - but I think you are on to something with your returns calculation. Also, if rents decline, expected income declines, lowering value. Owners don't like losing value. $\endgroup$
    – Jamzy
    Commented Sep 16, 2020 at 22:38

Whilst the other answers are very good I have identified a few drivers of this as well.

  • Residential rental yields have dropped to a very low rate. In Australia it is approx 3%, a retail property closer to 6%. In order to compete with other investments, these capital values need to increase and in Australia, this is what has been happening. If they didn't increase, the price couldn't be justified. Consequently, rental returns are a much smaller part of the equation. We have negative gearing in Australia, which exacerbates the situation by providing a subsidy on loss making investments.

  • Macrobusiness (paywall) has a different view and places the blame squarely on Chinese investors who have been far more dominant recently:

Chinese property investors are also renowned for keeping their homes vacant rather than renting them out.

'We know from Prosper’s Speculative Vacancies report that a significant percentage of homes in inner-Melbourne, where there is significant foreign investment in high-rise apartments, are being kept vacant.

Also in this article there is a reference to Australian property as a haven for laundered funds. I'd imagine that impacts appetite for risk and return requirements markedly.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.