Photo by Matthew Feeney on Unsplash

Cash flow corporate taxation (CFCT) has the potential to reduce the economic efficiency costs of the corporation tax, as Murphy (2018) has shown. This is because the tax applied to a marginal investment – one which yields only the risk-free interest rate, proxied by the bond rate – is zero.

The tax nonetheless generates revenue from more successful investments, ones that yield economic rent, and also has the potential to act as a lump sum equivalent tax on ‘old’ capital – that which was already in place at the time the tax was introduced. Garnaut et al (2018) have estimated that the revenue yield from a cash flow tax could better that of the existing corporate income tax.

Issues with various proposals for cash flow corporate tax

The classic form of a cash flow corporate tax was proposed by E C Brown in 1948. It allows immediate deductibility of capital expenditures (called expensing), but not for financing costs. For neutrality, it is a two-sided tax; that is, when cash flows are negative, the government pays a cash rebate calculated at the company tax rate. This makes the government an implicit partner in the business, participating in the success or otherwise of the business in proportion to the tax rate, without reducing the percentage yield for the investor. Meade (1978) called this the real cash flow corporate tax, or R-CFCT. The possibility of negative tax collections means that this tax has never been tried.

The resource rent tax proposed by Garnaut and Clunies Ross (1975) is a one-sided tax. Rather than government making payments when negative cash flows occur, the resource rent tax allows a tax rebate to be carried forward at an uplift rate. The risk is that the uplift rate may not sufficiently compensate for the riskiness of the investment, so that some may not proceed. Meade contemplated a similar mechanism for the R-CFCT. Such a tax is similar in concept to the allowance for corporate equity, whereby ‘normal’ returns are exempted.

The R-CFCT is not able to tax the financial sector. To address this, Meade proposed the R+F CFCT, which also included some financial flows so that interest, for example, could be an income or an expense (the F stands for financial flows). Meade noted that the R+F tax was identical to the S-CFCT, which is a tax on net flows to shareholders (the S stands for shareholders). This tax is extremely simple, as the base is dividends plus capital returns plus share buybacks, less new equity raisings. This can be understood as a lump sum equivalent tax on the corporation’s equity, as the value of that is the discounted value of all future cash flows to shareholders.

The S-CFCT is extremely simple but has several issues. One is the two-sided nature of the tax. Another is that it is not always easy to distinguish outgoings of an equity nature from those on loan account – for instance, due to hybrid debt instruments. A third is that the tax can be avoided by never paying out. For example, Berkshire Hathaway in the United States has never paid dividends. The government gains future revenue which may materialise only in the never-never. If this means that profits are re-invested, the outcome may be acceptable – it is a feature of this type of tax. But, if it simply reflects a build-up of cash, this is less acceptable.

The Z-tax and the S+L CFCT

The Z-tax seeks to address these concerns. Payouts to corporations in negative cash flow are avoided by giving a ZT credit, which is up-scaled each year by a variable that reflects the revenue need of the government. If the upscaling rate is the inflation rate, the tax becomes a sort of deferred corporate income tax. If it is the bond rate, the tax is closer in concept to the allowance for corporate equity. None of these uplift rates can aim to be fully neutral, as the literature on the resource rent tax makes clear. However, they are a lot more neutral than the corporate income tax. I call this system of uplifted tax credits the Z-tax corporate tax, or ZTCT.

To overcome the second issue, that of hybrid financing, we can expand the base of the S-tax to include all net payouts to loan providers. Murphy (2017) has shown that this approach is neutral, in the context of an economic rent tax (super profit tax) on the financial sector. It is a small further step to generalise this finding to make it into a general economic rent tax, which I call the S+L CFCT (where the L stands for loans). This converts the tax base into a lump sum equivalent tax on the whole of the enterprise value, rather than just the equity value. This is a very simple tax, as the base is net payments to owners and lenders. However, this disadvantages geared corporations at the time of the transition. To overcome this, ZT credits must be issued at that time proportionate to the corporations’ net debt.

To overcome the incentive for corporations to accumulate cash and not pay out, we can include net increase in cash in the tax base. To avoid double counting when the cash is eventually paid, tax payable on net increase in cash is accompanied by payment of a ZT credit. If cash has a net decrease, this reduces tax otherwise payable and reduces the ZT credit, so it is symmetrical.

An advantageous alternative

The S+L tax broadens the tax base compared to the S-CFCT as it includes any outward interest payments higher than the ZT uplift rate. My modelling shows that under some assumptions the tax rate can be less for a given revenue. This is important, as it is the headline tax rate which tends to drive avoidance behaviour such as offshore profit shifting. Note that such shifting is more difficult under the S+L base, as artificially high interest on related-party loans results in higher tax. However, shifting via inflated costs – for instance, intellectual property – is still possible. This is an inevitable drawback of a territorial (source based) tax system. A destination based cash flow tax might avoid this, but is not suited to taxing companies that export most of their product, like the mining sector.

A source-based tax on economic rents would allow the abolition of all specific mining royalties in Australia, with their associated high deadweight costs, and would also act as a super profit tax on very profitable sectors such as (at least until recently) banking.

The ZTCT could be introduced in conjunction with the current personal income tax, as the imputation system converts it into a withholding tax, and the economic characteristics of the system are then determined at the personal level. However, the tax is best conceived as an adjunct to a personal Z-tax, as described in my TTPI working paper 6/2019, or a personal direct consumption tax.


This article is based on a new Working Paper, Improving cash flow corporate taxation (CFCT) and the Z-tax (ZT) approach (paper no. 7/2019), from the Tax and Transfer Policy Institute, ANU.


Further reading

Issues with Capital Income Taxation and the Z-Tax Solution, by David Ingles (12 August 2019)

Leave a comment

Your email address will not be published. Required fields are marked *