I asked ChatGPT the following question:
What are the major effects of rent control on the quantity and quality of housing in the city where the rent control is imposed?
Here’s the answer I got back:
Rent control is a policy that limits the amount by which landlords can increase the rent for a rental property. The major effects of rent control on the quantity and quality of housing in a city depend on a variety of factors, including the specific details of the rent control policy, the local housing market, and the overall economic conditions of the city.
One potential effect of rent control is that it may reduce the quantity of rental housing available, as landlords may be less willing to invest in maintaining or improving their properties if they are unable to charge higher rents to cover the costs. This could lead to a decline in the overall quality of rental housing in the city. Rent control may also discourage new construction of rental housing, as developers may be less likely to build new properties if they are unable to charge market-level rents.
Except for the “mays,” this doesn’t seem too bad, right?
Unfortunately, I didn’t quote the whole thing. ChatGPT goes on to say:
On the other hand, rent control can also have positive effects on the quantity and quality of housing in a city. It can help to stabilize rents and prevent rapid rent increases, which can make housing more affordable for low-income residents and help to reduce the risk of homelessness. Rent control can also incentivize landlords to maintain and improve their properties, as they may be more likely to do so in order to retain their current tenants and avoid vacancies.
Overall, the effects of rent control on the quantity and quality of housing in a city can be complex and dependent on a variety of factors. It is important to carefully consider the potential impacts of rent control policies before implementing them.
“Rent control can also incentivize landlords to maintain and improve their properties, as they may be more likely to do so in order to retain their current tenants and avoid vacancies.” Seriously?
Keeping rents below where they would otherwise be gives landlords an incentive to maintain and improve their properties? The opposite is the truth. And they would have no trouble retaining current tenants and avoiding vacancies.
I see a number of my younger economics professor friends worrying on Facebook about how they’ll be able to grade students on written take home tests. Of course, one answer is: Don’t give written take home tests. But if this answer on rent control is any indicator, grading students shouldn’t be much harder than it currently is.
For a good piece on rent control, see Walter Block, “Rent Control,” in David R. Henderson, ed. The Concise Encyclopedia of Economics.
READER COMMENTS
John hare
Dec 20 2022 at 11:10am
I aught to write an article advocating wage control. It would keep job here locally and encourage more industries to move here. There would be job security and employers would be able to plan ahead so much better than the Wild West of wage competition.
Mark Barbieri
Dec 20 2022 at 12:52pm
It would actually be kind of funny to ask ChatGPT to write about wage control because it would have an incentive to discourage labor in favor of automation.
John hare
Dec 20 2022 at 1:06pm
I would like to see that. A strong article on wage control should bleed over to other controls in the reader’s mind. A majority of the voting public in some zip codes seem to think legislation overrides market forces.
Scott Sumner
Dec 20 2022 at 1:39pm
It will be amusing to see how students react when told that their ChatGTP answer is wrong.
robc
Dec 20 2022 at 2:49pm
Especially if told that specific way!
David Henderson
Dec 20 2022 at 5:25pm
Good point.
Kevin Dick
Dec 20 2022 at 2:48pm
I had an extended “dialogue” with ChatGPT about price discrimination.
It did a very good job of simulating a precocious high school student who wanted to parrot the textbook answer rather than explore the topic.
It initially painted itself into the corner that price discrimination only occurs when two people are charged different amounts for _exactly_ the same thing. I forced it to admit to specific cases where this wasn’t true. But it never gave up the underlying premise. In fact, it seemed to talk itself into an intrinsic theory of value that was distinct from customer preferences. It flat out stated that price discrimination did not occur when different willingness to pay was based on personal preference.
And it had issues with pricing that discrimination based on the time when the sale or use occurred and whether that reflected a personal preference.
Ahmed Fares
Dec 20 2022 at 11:17pm
One more problem that rent control causes. This from an article by Don Boudreaux:
Gavin R. Putland
Dec 21 2022 at 5:13am
What’s better than rent control? A tax on vacant lots and unoccupied buildings. While rent control makes it less attractive to supply accommodation, a vacant-property tax makes it less attractive NOT to! Such a tax, although sometimes called a “vacancy tax”, is not limited to what real-estate agents call “vacancies” — that is, properties available for rent. It also applies to vacant lots and empty properties that are not on the rental market, and prompts the owners to get them habitable and occupied in order to avoid the tax.
By the way, the desired *avoidance* of the vacant-property tax would increase economic activity, expanding the bases of other taxes and allowing their rates to be reduced, so that everyone else—including tenants, home owners, and landlords with tenants—would pay LESS tax!
Jon Murphy
Dec 21 2022 at 8:05am
I’m not so sure. A vacant property is vacant for a reason: the owner belives (rightfully or wrongfully) that it is better to remain vacant than take any offer they have. The size of the tax will naturally change that calculus.
But the calculus could change in unexpected ways. It’ll depend on what alternatives the owner is facing. For example, the owner may opt to change their building to commercial or retail space to avoid the tax. In that case, the tax would reduce the amount of supply on the market, not increase it.
My point is simply that a supposed market failure cannot be cleanly solved by a tax. We need to look at the relevant alternatives and how the tax will affect incentives given property rights.
David Henderson
Dec 21 2022 at 10:04am
Jon Murphy makes good points.
I want to add one that Jon hints at with his language “a supposed market failure.” It’s hard to see why there’s a market failure here. It seems that landlords would have the right incentive to trade off between having the apartment rented and leaving it vacant to find the right tenant or to make some improvements.
Also, it’s not clear that other people’s taxes would fall if this tax were imposed. You could check Vancouver, BC to see if this happened after the government imposed the tax you’re suggesting.
I do agree, though, that if the vacancy tax is small enough, it’s better than rent control because it doesn’t create non-market-clearing distortions.
BS
Dec 23 2022 at 1:19pm
Waiting to see how the proposed house flipping tax in BC will be charged as (if effected, in 2023) and whether it will militate against the vacancy tax by encouraging owners in some cases to keep vacant properties off the market for a few months…
Jose Pablo
Dec 23 2022 at 12:11pm
The owner of a vacant property is already paying a “tax” equivalent to the rent he is forfeiting for not renting the house.
It would be interesting to know the reasons why he/she is willing to voluntarily pay this “tax”. Acting upon these reasons would, very likely, be the best policy. Particularly if the reasons are “regulatory” or related to “property rights enforcement issues”, as it is frequently the case.
In any case an increase on market rents will play the same role that taxing “vacant properties”, so, the impact of vacant properties on market rents should be self-correcting.
And the resources wasted on enforcing and collecting this new tax could be used to do some productive useful work increasing the size of the economy.
Jim Glass
Dec 23 2022 at 1:15pm
Similarly, a tax on people who are unemployed makes it less attractive for them not to work — waste the productive capacity of their labor — while perhaps searching for the optimal job, making time-consuming hiring arrangements for a superior job, or waiting for market conditions to improve.
Make both these taxes BIG enough and you can fill all the vacant lots in Manhattan with fast food joints staffed by recent graduates of university humanities graduate degree programs.
Various
Dec 21 2022 at 10:12pm
That’s pretty funny. ChatGPT tried to put a good spin on both the pro and con arguments, even though they were in conflict with each other. At this rate, I’d say ChatGPT has excellent credentials to run for Congress! Basically fling the pasta on the wall and see what sticks.
Thomas Lee Hutcheson
Dec 22 2022 at 7:48am
It would be interesting to just ask it to perform the full cost-benefit analysis instead of just listing pros and cons. Or lay out how such an analysis would be performed.
David Henderson
Dec 22 2022 at 9:18am
The cost/benefit analysis is straightforward: with even one unit misallocated or less maintained, costs exceed benefits. Also, if rent control reduces quantity, costs exceed benefits. So in this case, you don’t need to ask for a cost/benefit analysis.
Notice also, Thomas, that I didn’t ask it for pros and cons. I asked it to state the effects of rent control on two things.
Jose Pablo
Dec 23 2022 at 1:16pm
Rent control is exactly the same that market rents + an (additional) tax on rental income (100% tax rate on rents above the stablished threshold) + a rent subsidy with similar scheme and with the same cost.
So, the benefits, if any, should be on the [an (additional) tax on rental income + a rent subsidy] part of the scheme.
I don’t see any, don’t we have enough redistribution schemes and dead weights? … I find it very unlikely giving the thousands of pages already in the tax code + another thousands regulating subsidies.
Not decreasing returns when it comes to tax code size and subsidies?
Jose Pablo
Dec 23 2022 at 12:16pm
It would be interesting to ask ChatGPT about the effects of increasing the minimum wages on the labor market.
Is ChatGPT expecting to write interesting papers (instead of helping students to provide doubtful answers) any time soon?
Maybe, teachers should be worried about ChatGPT replacing them, not about students using ChatGPT to cheat … First till year 8, then in high school and later on …
Mauro Mello Jr
Jan 7 2023 at 4:19am
[Only found this article today, Sat 2023-01-07…]
You can use the GPT-2 Output Detector (https://openai-openai-detector.hf.space/) to find out the probability that the text is fake (that is, machine learning-generated) or not.
Either singly or in combination, both excerpts of the answer you got above result in 99.98% prediction that the text is fake. (And you do not need this test to find out that the second excerpt could not have been generated by a thinking economist.)
Don Boudreaux’s text (see entry by Ahmed Fares above (2022-12-20) gets a 99.97% real. Don’s clean.
The first four paragraphs in Walter Block’s text on rent control that you linked to get a 99.96% real. Block’s clean, too.
(This article on Medium by Sung Kim, How to Detect OpenAI’s ChatGPT Output (https://medium.com/geekculture/how-to-detect-if-an-essay-was-generated-by-openais-chatgpt-58bb8adc8461), has references and more resources.)
Comments are closed.