I can imagine that if the stamp duty to LVT switch increased the tax concessions for owner-occupiers relative to investors, then the composition of the seller and buyer cohorts would change in a way that produced the result. But then the result wouldn't be due to the removal of the transaction tax per se. We could keep stamp duty but make it punitive for investors and it would achieve the same thing.
From your reading, however, that's not why the result occurred. The result was by assumption. Interesting!
I have reached out to the author to ask him to upload a non-paywalled version.
Sep 22, 2023·edited Sep 22, 2023Liked by Cameron Murray
It becomes increasingly obvious from Cam's arguments with property tax reform that specially targeted tax design makes more sense for achieving specific objectives rather than wholesale broad based tax changes for the sake of neutral tax treatment and "efficiency". Especially given the high political (and/or fiscal) cost of making these tax changes.
Do we want more new dwellings? Then tax new dwellings less.
Do we want more homeowners? Then tax people who don't own a home less. Tax people who own more than 1 property more.
Do we want wealthy people to stop banking wealth in their PPR as a tax shelter and keeping potentially good development sites off the market? Then tax those properties more.
Do we want pensioners to not face barriers to downsizing due to asset testing? Then create tenure neutrality for asset testing the pension.
Maybe if we actually started talking about making small tweaks that moved in these directions and made it a long term objective, we would start getting somewhere.
How about eliminate the stamp duty tax for owner-occupants and double it for landlords and see what happens to home ownership over time? Or double the LVT for landlords and guess what will happen to home ownership.
If you want less of something (landlord-ownership), increase the tax. If you want more of something (home-ownership), decrease the tax.
I agree it's hard to understand how they got to those numbers. However, I think there are other good reasons to abolish stamp duty. First it allows greater movement to upsize / downsize as required, which has a flow on effect to the community infrastructure required (in particular schools, playing fields etc). Secondly, my observation is that stamp duty impacts lower income families more frequently - this is because they are often in jobs which are decentralised (teachers, nurses, tradies) and therefore they either spend hours in a car to get a promotion in a school on the other side of the metro region, or need to move. Compare this with people whose jobs will always be locationally static and metro-centric, who might not need to move for decades. That seems unfair to me.
On balance I think you make a reasonable point about residential stamp duty. It's more likely the NSW stuff was going to increase homeownership through higher taxes on investors resulting in them being less competitive in the market.
Really the only effective approach to having lower (or no) taxes is for new dwellings only, to increase demand for new dwellings which results in more supply. The secondary market as you point out is less of a concern.
There is a point though that there would be an optimal level of stamp duty, or perhaps a laffer curve. At some point the MEB would get highly damaging even if the Average EB is low.
Additionally, I think the downsizing issue for stamp duty does exist but it's nowhere near as bad as asset testing in the pension. We really need to do something about excessive wealth sheltering in owner-occupied housing. Some sheltering is benign, but the excessive levels of millions of dollars in well located mansions and enabling people to turn potential development sites into personal tax sheltered estates is both bad for equity and efficiency.
Capping the PPR land tax exemption to a certain amount of land value and/or tweaking asset testing would go a long way to addressing this.
Setting it to only affect say the top 1% of properties by land value and then allowing bracket creep to ease in the transition could be a plausible method.
Interesting read and very clear argument. Beyond just renters, I think this hurts the experience of home owners. My neighbour told me that the house next door being bought by a landlord was one of the worst things that happened to the street. After that they had new neighbours every 1-2 years, and he told me he started caring less about the neighbourhood. His kids made friends with the first kids next door, but not anymore after that. It's anecdotal, and just a 'street' of just 7 houses, but sounds plausible as a general pattern. It's a cost that is largely invisible to the government spreadsheets.
Has anyone looked into the cost of increased transactions? Higher car speed and faster patient turn around has been linked to an increase need for roads and hospital beds, because of 'buffers' required between cars/patients. I expect this to be even more true for houses. A house will easily be empty for 2 weeks (and/or people renting two places). With some renters moving almost yearly, if that experience would be universal, that would add up to 3-4% extra housing demands, which seems significant.
Good post. I struggled to understand how they assumed greater selling meant 6% increase in ownership. As you say, you’d need both volume increase in people selling but for those selling to also be selling (permanently) their investment properties AND have them sold/bought by owner occupiers. Ie like you say you need concentration to drop. I think the NZ case is intriguing but you also need to consider all the other factors around homeownership in NZ over the same time period. I didn’t read anything you covered in that regard - NZ also went on a massive social housing build that would’ve timed up with that period post stamp duty removal too and because of their quality reduced the need for some to have a desire to own. There’s a host of other reasons including tax settings that favour property investors vs owner occupiers and costs to build in our far away country making buying hard but it’s far too simplistic to say stamp duty removal would fail because hey look it did in NZ. I personally don’t like the concept of taxes on a transaction the government did nothing in relation to - a broad capital gains or tax on land discriminated less and removed the need for stamp duty to push up transaction costs (which are often the difficulty of affording a house vs ongoing servicing). I think the conversation about removing stamp duty is also useless without also discussing how you’d offset it with other revenue or decreased spending.
Taxing transactions rather than final consumption (or imputed income if "consumption" is not feasible) seem like a bad idea whether applied to loaves of bread, shares of stock, or real estate. Possibly mistaken claims of benefits aside, what is wrong with substituting a land value tax (or even a tax on total value) for a transaction tax?
Thanks for diving into the stamp duty paper.
I can imagine that if the stamp duty to LVT switch increased the tax concessions for owner-occupiers relative to investors, then the composition of the seller and buyer cohorts would change in a way that produced the result. But then the result wouldn't be due to the removal of the transaction tax per se. We could keep stamp duty but make it punitive for investors and it would achieve the same thing.
From your reading, however, that's not why the result occurred. The result was by assumption. Interesting!
I have reached out to the author to ask him to upload a non-paywalled version.
It becomes increasingly obvious from Cam's arguments with property tax reform that specially targeted tax design makes more sense for achieving specific objectives rather than wholesale broad based tax changes for the sake of neutral tax treatment and "efficiency". Especially given the high political (and/or fiscal) cost of making these tax changes.
Do we want more new dwellings? Then tax new dwellings less.
Do we want more homeowners? Then tax people who don't own a home less. Tax people who own more than 1 property more.
Do we want wealthy people to stop banking wealth in their PPR as a tax shelter and keeping potentially good development sites off the market? Then tax those properties more.
Do we want pensioners to not face barriers to downsizing due to asset testing? Then create tenure neutrality for asset testing the pension.
Maybe if we actually started talking about making small tweaks that moved in these directions and made it a long term objective, we would start getting somewhere.
How about eliminate the stamp duty tax for owner-occupants and double it for landlords and see what happens to home ownership over time? Or double the LVT for landlords and guess what will happen to home ownership.
If you want less of something (landlord-ownership), increase the tax. If you want more of something (home-ownership), decrease the tax.
What if instead of resulting in less concentration, it just resulted in faster or slower more?
I agree it's hard to understand how they got to those numbers. However, I think there are other good reasons to abolish stamp duty. First it allows greater movement to upsize / downsize as required, which has a flow on effect to the community infrastructure required (in particular schools, playing fields etc). Secondly, my observation is that stamp duty impacts lower income families more frequently - this is because they are often in jobs which are decentralised (teachers, nurses, tradies) and therefore they either spend hours in a car to get a promotion in a school on the other side of the metro region, or need to move. Compare this with people whose jobs will always be locationally static and metro-centric, who might not need to move for decades. That seems unfair to me.
On balance I think you make a reasonable point about residential stamp duty. It's more likely the NSW stuff was going to increase homeownership through higher taxes on investors resulting in them being less competitive in the market.
Really the only effective approach to having lower (or no) taxes is for new dwellings only, to increase demand for new dwellings which results in more supply. The secondary market as you point out is less of a concern.
There is a point though that there would be an optimal level of stamp duty, or perhaps a laffer curve. At some point the MEB would get highly damaging even if the Average EB is low.
Additionally, I think the downsizing issue for stamp duty does exist but it's nowhere near as bad as asset testing in the pension. We really need to do something about excessive wealth sheltering in owner-occupied housing. Some sheltering is benign, but the excessive levels of millions of dollars in well located mansions and enabling people to turn potential development sites into personal tax sheltered estates is both bad for equity and efficiency.
Capping the PPR land tax exemption to a certain amount of land value and/or tweaking asset testing would go a long way to addressing this.
Setting it to only affect say the top 1% of properties by land value and then allowing bracket creep to ease in the transition could be a plausible method.
I can accept your argument but it would benefit from another case study beyond NZ, so it does not appear to be only anecdotal.
Canberra is another case to look at. Let me see if there is any evidence there about lower stamps duties and higher LVT increasing homeownership
Interesting read and very clear argument. Beyond just renters, I think this hurts the experience of home owners. My neighbour told me that the house next door being bought by a landlord was one of the worst things that happened to the street. After that they had new neighbours every 1-2 years, and he told me he started caring less about the neighbourhood. His kids made friends with the first kids next door, but not anymore after that. It's anecdotal, and just a 'street' of just 7 houses, but sounds plausible as a general pattern. It's a cost that is largely invisible to the government spreadsheets.
Has anyone looked into the cost of increased transactions? Higher car speed and faster patient turn around has been linked to an increase need for roads and hospital beds, because of 'buffers' required between cars/patients. I expect this to be even more true for houses. A house will easily be empty for 2 weeks (and/or people renting two places). With some renters moving almost yearly, if that experience would be universal, that would add up to 3-4% extra housing demands, which seems significant.
Good post. I struggled to understand how they assumed greater selling meant 6% increase in ownership. As you say, you’d need both volume increase in people selling but for those selling to also be selling (permanently) their investment properties AND have them sold/bought by owner occupiers. Ie like you say you need concentration to drop. I think the NZ case is intriguing but you also need to consider all the other factors around homeownership in NZ over the same time period. I didn’t read anything you covered in that regard - NZ also went on a massive social housing build that would’ve timed up with that period post stamp duty removal too and because of their quality reduced the need for some to have a desire to own. There’s a host of other reasons including tax settings that favour property investors vs owner occupiers and costs to build in our far away country making buying hard but it’s far too simplistic to say stamp duty removal would fail because hey look it did in NZ. I personally don’t like the concept of taxes on a transaction the government did nothing in relation to - a broad capital gains or tax on land discriminated less and removed the need for stamp duty to push up transaction costs (which are often the difficulty of affording a house vs ongoing servicing). I think the conversation about removing stamp duty is also useless without also discussing how you’d offset it with other revenue or decreased spending.
Taxing transactions rather than final consumption (or imputed income if "consumption" is not feasible) seem like a bad idea whether applied to loaves of bread, shares of stock, or real estate. Possibly mistaken claims of benefits aside, what is wrong with substituting a land value tax (or even a tax on total value) for a transaction tax?