The tax/rent trade-off

Property investors, when objecting to taxes on their rents and capital gains, profess to be providing for their retirements through the property market in order to avoid burdening the taxpayers. This, they insinuate, is as good as paying tax. So where is the justice in making them pay tax too? And what is the point of taxing them if it only causes a drain on social security funded by other taxpayers?

In response one might point out — if one were allowed to — that property investors receive their rents at the expense of tenants and receive their capital gains at the expense of first home buyers. Is there any good reason why a retiree funded by tenants and first-time buyers should be considered more respectable than one funded by taxpayers?

Yes, there is. Retirees funded by tenants and first home buyers do not damage the economy, but are merely the receivers of natural market prices, whereas taxes deter productive activity and consequently make the nation poorer.

But in that case, should not the government imitate the property investors, making itself a receiver of natural market prices in lieu of taxes? Moreover, if the received market prices only replace the revenue from the taxes, should not the right to receive those prices have the same market value as the right to impose the taxes, in which case the government, without resorting to any form of compulsion, could buy the former right by surrendering the latter? We answer in the affirmative.

2. Theory

2.1 All taxes fall on rent.

Suppose that all individuals and firms located entirely within a certain zone were exempt from federal taxes [1]. The exemption would induce a net migration of taxable entities (individuals and firms) into the zone, driving up the rents (hence prices) of land in the zone until the rent premium (or interest premium) cancelled the tax advantage, at which point the net migration would cease. This "rent premium" is the difference between the rent of land inside the zone and the rent of comparable land outside the zone. If the zone were very small, the effect of the migration on land rents outside the zone would be negligible, so that the rent premium would be almost entirely manifested as a rise in rent within the zone. Moreover, even if the zone were larger or of no fixed size, the overall supply of land is still fixed, so that any increase in effective demand for land cannot be offset by increased supply; therefore the increase in spending power conferred by the exemption would tend to be competed away in the land market, so that the rise in rent inside the zone would not be balanced by a fall in rent outside the zone.

It follows (at least to a first approximation) that the federal taxes paid by tenants are simply deductions from the rent that the landlords could otherwise get for the land.

2.2 All deadweight costs fall on rent.

In this age of "self-assessment", most taxes impose substantial compliance burdens on the entities that pay and/or collect the tax. Both the taxes themselves and their compliance burdens reduce the returns on various economic activities, causing some otherwise viable activities to become unviable. The net cost of these lost opportunities is the so-called deadweight cost of taxation.

Under the above exemption, the "tax advantage" of being located in the designated zone is not limited to the actual tax saved, but also includes the additional opportunities opened up by the lifting of the tax burden; that is, it includes the removal of the deadweight cost.

It follows (again to a first approximation) that the deadweight costs of federal taxes on tenants are simply deductions from the rent that the landlords could otherwise get for the land.

(Note: From the viewpoint of a single taxpaying entity, compliance costs are uncompensated losses. But from the viewpoint of society, compliance costs incurred in cash are part of the income of other entities and are therefore not a net loss (although they still cause a net loss). For this reason, not all compliance burdens can be classified as deadweight, although all are causes of deadweight. In this paper we treat deadweight as being net of compliance burdens, so that, from the viewpoint of a single entity, deadweight costs and compliance costs are additive.)

2.3 Less tax: more revenue?

We have seen that not only the federal taxes paid by tenants, but also their deadweight costs, are deductions from the rent that the landlords could otherwise get. In the case of owner-occupants, who are both tenants and landlords, the implication is that the taxes and their deadweight costs are deductions from the imputed rent of the land; but this is even more obvious, as imputed rent is simply the benefit of owner-occupancy. In general, then, both the taxes and their deadweight costs are deductions from the rental value of the land, whether the land is actually "rented" or not.

If the taxes were avoided, the rental value of land would increase by a margin equal to the avoided taxes plus the avoided deadweight costs, i.e. by more than the avoided taxes. Hence, if this increase in rent could be captured for public revenue, it would more than compensate for the avoided taxes while leaving the land owners no worse off. The extra revenue could be given back in the form of further tax cuts or spent on additional public services.

3. Implementation

3.1 An optional tax/rent swap

Let f be the estimated total revenue from recurrent income tax, GST and sales taxes, expressed as a fraction of the total rental value of land in the Commonwealth. Because of the deadweight effect, abolition of those taxes would increase the rental value of land by a somewhat larger fraction. The Commonwealth might try to capture the rent rise by making the following offer to every taxpayer: if you pay us a fraction f of the rental value of all land that you own or rent, you are exempt from income tax on recurrent income (including rental income) and from GST and sales taxes. To allow for variation in the circumstances of taxpayers, the Commonwealth might also pay a negotiable lump-sum P to each taxpayer who enters the new system; the value of P (meaning principal or price) could if necessary be negative, in which case the payment would be in the opposite direction. We shall now show, by means of a little algebra, that there is a range of values of P for which the proposed deal between the Commonwealth and the taxpayer ("you") would be mutually beneficial.

Let R be the total rental value of all the land that you own or rent (which would be known from, e.g., valuations for municipal rates and State land tax). Let T be the recurrent tax bill that you stand to avoid, C the recurrent compliance cost that you stand to avoid, D the recurrent deadweight cost that you stand to avoid, and i the discounting rate (real interest rate).

Then the deal is attractive to you provided that the benefit exceeds the cost — i.e. provided that

where iP is the annualized equivalent of the lump-sum payment (i.e. the real interest earned or saved thereon).

where, again, the benefits are written on the left and the costs on the right, but from the Commonwealth's point of view.

The two conditions are compatible because C and D are missing from (2). In practical terms, that means they are compatible because the deal gets rid of your compliance cost and deadweight cost, leaving a net benefit to be divided between you and the Commonwealth.

where the middle expression (fR−T−iP) is recognizable as the net benefit to the Commonwealth.

Now notice that for any value of f, there is a range of values of P that satisfies (3). At the minimum value of P, condition (3) becomes

so that the Commonwealth gets the whole benefit of the removal of compliance costs and deadweight costs (C+D). At the maximum value of P, condition (3) becomes

so that the Commonwealth gets none of the benefit (i.e., the taxpayer gets it all). So the lump-sum will be negotiated to give a mutually acceptable division of the benefit, and the deal will go ahead.

3.2 No fiscal deficit

The Commonwealth's share of the benefit means that each deal will be revenue-positive. That guarantees that there will be no "black hole" in the federal budget in consequence of this reform. A separate macroeconomic calculation demonstrating that the budget figures "add up" is neither necessary nor possible, because all the "adding up" is done on the microeconomic scale, one deal at a time.

Recall, moreover, that the above reasoning holds for any value of f. That is why we initially defined f as the estimated total revenue from the nominated taxes, expressed as a fraction of the total rental value of land. The balance of the federal budget does not depend on the accuracy of this "estimate". If the estimate is accurate, then on average fR will cancel T in (3), so that the average mutually acceptable value of P will be slightly negative. If the estimate is bad, the worst-case scenario is that the average value of P will be large, causing a substantial change in the public-sector borrowing requirement offset by an equal and opposite change in the private-sector borrowing requirement. In that case, the planned tax/rent trade-off would become, on average, a tax-rent-interest trade-off; but, as there would be no increase in the overall borrowing requirement, there would be no consequent increase in interest rates.

As the estimated total rental value of land in the Commonwealth would be based on the same valuations as the total rental values for individual taxpayers, those valuations are the most likely source of error. But in that case, there will be a compensating error in the calculated value of f and no effect on the total or average value of P. So the "worst case", which is not too bad, is also not too likely.

3.3 Coexistence of two systems

Entities under the new system — that is, entities that have accepted tax/rent swaps — would not pay income tax on recurrent income, and would not remit GST (VAT) or sales taxes. But they would still bear any indirect taxes hidden in prices that they pay (although those hidden taxes would be reduced as other entities gained exemptions from remitting indirect taxes), and would still remit personal income tax on behalf of any employees who are not themselves exempt [2].

Consider GST. A business under the new system would not remit GST on its sales. And as it would "bear any indirect taxes hidden in prices", it would not claim GST credits on its inputs; this conclusion is also consistent with the aim of eliminating compliance costs. In summary, the business would be GST-exempt in the internationally accepted terminology, or input-taxed in the Australian terminology. Widespread acceptance of tax/rent swaps would simply cause a proliferation of input-taxed businesses, which other businesses would deal with according to the existing rules. If this in turn causes political pressure to simplify the rules, so be it [3].

The reason why the interaction between an entity under the new system and an entity under the old system raises a question in connection with GST is that the existing GST rules raise the same question. But the same existing rules also answer the question. If the interaction between the two systems also raises issues concerning income tax or sales taxes, the same logic should apply; the coexistence of the old tax system and the new rent system should not cause any new problems.

3.4 Scope of "land"

The market value of a piece of "land" includes the value of any building rights conferred by the planning or zoning system, and of any other privileges attached to the "land" or building(s), because all these things affect the potential rent stream. To allow for this, the category "land" must be generalized to include sites — that is, pieces of ground or airspace, including any attached rights to build on that ground or into that airspace or to use the ground or buildings for particular purposes, but excluding any actual buildings.

When we say that the rise in spending power conferred by a tax exemption would be converted into a rise in the rental value of land, it is critical that taxpayers cannot produce more land or bring more land into the taxing jurisdiction; if they could, the additional supply of land would restrain the increase in rent. This indeed is the essential economic property of sites: taxpayers can neither create sites nor move them into (or out of) the taxing jurisdiction. But sites are not the only assets with this property.

Consider taxi licences, also known as plates. In Australia, taxi plates are issued by State governments and are usable only within their respective States (or specified parts thereof). They cannot be created by taxpayers or moved between States, let alone moved into or out of Australia. So for economic purposes they are site-like. For a taxi operator per se, the one indispensable site-like asset is not a site, but a plate. Hence any tax advantage conferred on taxi operators would tend to be competed away, not in the rental values of sites, but rather in the rental values of plates. If plates were not assessable in the tax/rent trade-off, taxpayers in other lines of business could opt out of the present tax system and then switch to the taxi business, raising the non-assessable rents of plates and lowering the assessable rents of sites, causing a net loss of revenue.

So, if the thesis that all taxes and all deadweight costs fall on the rent of "land" is to be strictly true, and if leakages of revenue are to be prevented, the category "land" must be further generalized to include all site-like assets — that is, assets that taxpayers can neither create nor move into (or out of) the taxing jurisdiction.

Site-like assets include not only sites and taxi plates, but also gaming licenses, electromagnetic spectrum assignments, airport and seaport time slots and other rights of way, pollution rights, water rights, fishing rights, forestry rights, patents, and copyrights. Of these, all except patents and copyrights have observable rental values (or at least lump-sum values from which equivalent rental values can be calculated) and therefore can easily be counted as "land" whose rental value is assessed for the purpose of tax/rent swaps.

To the difficulty caused by patents and copyrights, the obvious solution is to exclude these assets from assessable "land", and likewise to exclude the associated royalties from the category of "recurrent income" that is exempt from income tax — in other words, to leave intellectual property under the old tax system.

Corporate shares are partly site-like in the sense that they are partly backed by other site-like assets, and in the sense that their supply is inelastic in the short term. Shares should therefore be assessed as "land", but using a reduced interest rate for the conversion of the lump-sum value to the rental value. As long as corporate profits are taxed in the hands of shareholders, the shareholders must be treated as the "land" owners for the purpose of tax/rent swaps. But if all taxation of listed companies were at source, the shareholders would not need to be treated as the owners, and shareholders joining the new system would consequently avoid compliance costs associated with buying and selling of shares and with fluctuating share values. If this causes political pressure from shareholders to reform the existing income tax so that corporate profits are taxed at source, so be it.

Of course, a listed company itself would be free to accept a tax/rent swap. To prevent any leakage of revenue, the geographic spread of assessable land-like assets would need to coincide with the geographic spread of the Commonwealth taxing power. In the case of a multinational company, this requirement could be so complicated as to prevent agreement. But if all or part of the multinational's operations within the Commonwealth were devolved to a separate company, that company could negotiate a tax/rent swap without difficulty. If this opportunity encourages a voluntary break-up of multinational companies into smaller companies with stronger national loyalties, so be it.

But most taxpayers, especially individuals and small businesses, would not be affected by "site-like" assets other than sites.

3.5 Scope of "recurrent income"

Because a rent or "rental value" is by definition an annuity, a tax/rent swap is necessarily restricted to recurrent taxes. It can include a negotiated lump-sum payable at the time of the swap; but it cannot include one-off tax liabilities associated with future capital transactions, because these are not necessarily predictable, and because intentions concerning them can be misrepresented.

Accordingly, the income tax exemption conferred by a tax/rent swap is restricted to tax on recurrent income; capital gains tax (CGT) is untouched. This does not mean that CGT cannot be reduced or abolished at some future time; but it does mean that CGT, by its very nature, cannot be included in the initial tax/rent swap.

Furthermore, if it is impractical to assess the rental value of certain site-like assets as part of a tax/rent swap, the recurrent income from those assets must be excluded from the income tax exemption in order to prevent a leakage of revenue; in other words, the present taxation of that income must continue. This has already been explained in connection with patents and copyrights. Again, this does not mean that the taxation of such income cannot be reformed at some future time; it simply means that such reform cannot be part of the initial tax/rent swap.

3.6 Scope of "sales taxes"

In Australia, following the introduction of the GST, the only remaining "sales taxes" are retail taxes on "luxury goods". While these taxes are ostensibly imposed in the name of equity, their effect is to increase the resale values of the taxed goods for the benefit of people who already own them, while pricing them further out of the reach of people who want them but cannot yet afford them. Accordingly, we submit that such taxes should be eliminated in any tax/rent swap.

It might be argued that "sales taxes" should include sumptuary taxes — also called "sin taxes" — which are intended (or so it is said) primarily to discourage undesirable behaviour, and only incidentally to raise revenue. Opinion on sin taxes, even within the membership of Prosper Australia (publisher of this paper), is far from unanimous. Some say that governments should not profit from activities that they ostensibly want to discourage, in which case the discouragement is better achieved by propaganda than by taxation. To the contrary, some argue that a tax is a fine by another name, and that it is better to fine people for undesirable behaviour than for desirable behaviour. (One might also point out that the revenue raised through sin taxes is at least not raised from illegal activity, which is more than can be said for the revenue raised through what are usually called fines.) Between these extremes, some argue that the purpose of sin taxes is to compensate the community for the costs attributable to the "sin", in which case the taxes should be high enough to cover those costs, but no higher. It is clear, however, than any exemption from sumptuary taxes would be controversial and would be widely perceived as socially and environmentally irresponsible.

We therefore suggest that in this paper, "sales taxes" should be taken to exclude sumptuary taxes such as excises on tobacco, liquor, and fossil fuels. Under this interpretation, all businesses that presently remit sumptuary taxes would continue to do so, while their customers would continue to pay sumptuary taxes hidden in prices. We say this not by way of endorsement of the present sumptuary tax regime, but by way of declaration that, for better or worse, reform of sumptuary taxes is beyond the scope of this paper.

While the retention of sumptuary taxes would cause compliance burdens for the businesses that actually remit the taxes, the number of affected businesses would be small as long as the taxes are remitted at wholesale level or further upstream. Even those businesses, by accepting tax/rent swaps, would be able to rid themselves of the compliance costs and deadweight costs associated with various other taxes. So the popularity of the tax/rent swaps would not be greatly affected.

4. Economic benefits

4.1 Advantages for enterprises

If your enterprise accepts a tax/rent swap, it avoids substantial compliance costs and deadweight costs. Most taxes and some compliance costs are marginal costs (that is, they increase with turnover), and therefore must be recovered through prices if your business is to grow. This is obviously an impediment to growth — that is, a cause of deadweight. In contrast, rent is a fixed cost (that is, independent of turnover). If you accept a tax/rent swap, thus replacing a marginal cost (tax) with a fixed cost (rent), you gain an advantage in the contest for market share.

4.2 Advantages for individuals

If you accept a tax/rent swap as an individual, your employer need not deduct income tax on your behalf, and can more easily reward you for good performance because income tax will not eat into any raise in your wages or salary. Thus you minimize compliance costs for prospective employers, increase your opportunities to improve your lot by hard work, and send a signal to prospective employers that you are keen to maximize your performance.

4.3 Lower unemployment without inflation

GST and sales taxes obviously feed into prices. Income tax, although usually called a direct tax, is a cost of production, and will prevent production unless it is recovered through prices. So all these taxes raise prices, thus increasing inflationary pressures and raising the so-called natural rate of unemployment, which is the minimum unemployment rate that causes enough wage restraint to give stable inflation. The central bank adjusts interest rates so as to maintain unemployment at this "natural rate".

But, as more and more entities accepted tax/rent swaps, fewer and fewer would remain subject to the inflationary taxes; so inflationary pressures, and therefore the "natural rate" of unemployment, would fall.

4.4 Trade surplus; falling foreign debt

When taxes feed into prices, they feed into prices of exports and import replacements and thereby damage international competitiveness. While GST purportedly does not apply to exports, the related compliance costs still affect export prices. Moreover, GST and sales taxes, by raising the cost of living, affect wage outcomes, hence export prices.

The tax/rent swaps, by removing a range of taxes that feed into prices, would improve the competitiveness of the nation's products, allowing the nation to trade its way out of debt.

4.5 Benefits for property owners through infrastructure

The fractional rent received by the Commonwealth after the tax/rent swaps would give the federal government an incentive to do things that increase the rental value of land — such as spending money on infrastructure. But remember that the rent received by the Commonwealth would not be at the expense of land owners; it would be additional rent generated by the elimination of selected taxes. Because of the fixed proportionality between the land owners' rent and the Commonwealth's rent, the infrastructure would increase both, delivering a benefit to the land owners. In most cases this benefit would not be delivered under the present system, because the infrastructure would not be built!

4.6 Bonanza for accountants, realtors, conveyancers

Millions of entities contemplating radical optional rearrangements of their tax affairs would obviously generate much business for the professionals who facilitate such rearrangements or give advice thereon. Property owners who accepted tax/rent swaps would see property in a new light — not only as a source of capital gains and (real or imputed) rent, both augmented by public expenditure on infrastructure, but also as a holding cost (the fractional rent bill accepted in lieu of the tax bill). Hence, while increased expenditure on infrastructure would make property investment more profitable, investors wishing to take maximum advantage would need to rationalize their holdings, either generating income from their properties or selling them to others who would generate income from them. That would generate much business for the property transaction industry.

Property sales between entities under the new system would cause no leakage of revenue because the fractional rent liabilities would simply be transferred from the sellers to the buyers. But the sale of a vacant property by a seller under the new system to a buyer under the old system would tend to reduce the public revenue received from that property, and the non-performing property would also have helped the seller to obtain a more generous lump-sum in the transition. To prevent such losses of revenue, the Commonwealth would obviously refuse to sign tax/rent deals with taxpayers carrying idle assets until the assets were either fully employed or disposed of. But that reinforces the point that property owners would gain an incentive either to generate income from their properties or to sell them.

4.7 Affordable accommodation

While rents received by landlords after the tax/rent trade-off would increase due to improved infrastructure and faster economic growth, this does not mean that accommodation would be less affordable, because affordability depends not only on the rent or price, but also on the amenity of the accommodation, which is improved by infrastructure, and on the spending power of tenants and potential buyers, which is improved by economic growth. To see the combined effect on affordability, one must consider the competitive position of tenants and potential buyers. By encouraging landlords to generate income from their properties, the tax/rent swaps would stimulate construction of new accommodation and make landlords less tolerant of vacancies. The improved supply of accommodation would reduce the intensity of competition between tenants and between potential buyers. On balance, then, accommodation would become more affordable, although rents and prices would increase in terms of raw dollars. To claim that both landlords and tenants would gain, or that both buyers and sellers would gain, is not a contradiction, because the promotion of economic growth is not a zero-sum game; it is analogous to making a bigger cake so that everyone can have a bigger slice.

5. Enabling legislation

As an educational association, Prosper Australia does not endorse any particular legislative plan. But this lack of endorsement does not imply lack of discussion, because Prosper Australia specializes in public finance, and no system of public finance is feasible unless it can be codified in legislation. Furthermore, no legislation can be passed unless it is politically acceptable.

The legislative embodiment of our "tax/rent trade-off" would be unusually simple. The essence of it would be contained in a single sentence such as the following:

Of course some definitions would also be needed. The types of "assessable assets" would need to be enumerated; these would be "site-like" assets other than patents and copyrights. The authorities for asset valuations and (where necessary) the discounting rates for converting capitalized asset values to rental values would need to be specified. The definitions of "recurrent income" and "luxury goods taxes" would presumably be taken from existing legislation.

The optionality of the tax/rent swaps has two far-reaching implications for the enabling legislation:

5.1 Political feasibility

Any real or imagined objection to the tax/rent trade-off is adequately answered by the fact that the trade-off would be optional. No one would be harmed by having the choice of staying in the present tax system or escaping from it. Any opponents of the reform would have to explain why they want to impose the status quo and deprive the people of choice. That could be difficult.

6. Conclusion

If a taxpaying entity could avoid recurrent income tax, GST, and sales taxes by paying a certain fraction of the rental values of land-like assets that it owns or rents, and paying or receiving a negotiated transitional lump-sum, both taxpayers and the Commonwealth would be keen to take up the offer. The damaging effects of taxation on the economy would fall in synchronism with the number of entities remaining in the old tax system. Thereafter, the government would fund a large part of its expenditure not from taxation, but from rent streams, which the government would effectively have purchased by giving up some of its taxing powers.

Notes

[1] Under s.51(ii) and s.99 of the Australian Constitution, such a zone could not be established within a State, but could be established within a Territory.

[2] If personal income tax were deducted and remitted by the financial institutions into which wages and salaries are paid, all employers would be relieved of this task in respect of all employees, with or without any tax/rent swaps. Similarly, compulsory superannuation contributions could be rolled into gross wages and salaries and then deducted by the same financial institutions, relieving employers of the task. Cf. Gavin R. Putland, "Give banks the job of deducting tax, super", letter to The Age, March 29, 2006, p.14.

[3] In Australia, a GST-registered business must treat an input-taxed (i.e. GST-exempt) supplier as if that supplier were zero-rated; that is, input credits are not claimable on inputs purchased from the input-taxed supplier even though there is GST embedded in the prices of those inputs. This is double taxation, and causes an exception in the procedure for handling inputs. Neither the double taxation nor the exception occurs for fully taxable suppliers, and the double taxation does not occur for zero-rated suppliers. For these reasons, many GST-registered businesses simply refuse to deal with input-taxed suppliers. In response, hundreds of thousands of small businesses and charities that would otherwise have been eligible for input-taxed status have been forced to register for GST and incur compliance costs. The problem would be avoided if input credits were claimable on inputs purchased from input-taxed suppliers. This would mean that GST is reclaimed on the output prices of an input-taxed business when in fact it has only been paid on the input prices, so that the value added by that business escapes taxation. Logically, that is exactly what should happen to a business that is "exempt" from a "value-added" tax!

Revision history

Version 2006.06.21 was the original, but drew heavily on Working Paper No. 1.

Version 2006.06.22:

Changed the title of subsection 3.3 and made its content more specific;

Noted the implications of optionality for the enabling legislation;

Made a CORRECTION to subsection 4.6 concerning the effects of property sales on revenue.

Version 2006.08.24 made a few textual corrections and added a link to the collected Working Papers.

Version numbers are dates in the form yyyy.mm.dd. Minor textual corrections and adjustments are not necessarily recorded.

The tax/rent trade-off:
Making tax (literally) optional

Contents

1. An unheard debate