Inflation and the tax system

When I went looking for the interim report of the Tax Working Group, I found that various other papers had been released.   These include background papers prepared by the Treasury and IRD secretariat looking at various possible options for reducing other taxes if, for example, new capital taxes were to provide more government revenue.

Among them was a short and rather unconvincing paper on productivity.   It was notable for highlighting how difficult it was to give any concrete meaning to the aspiration repeatedly expressed by the Minister of Finance, and included in the terms of reference, of “promoting the right balance between the productive and speculative economies”.  And it was also notable for the aversion of officials to lowering the company tax rate (or the effective tax rate shareholders pay on company income), even though they accept that our business income tax rates are now high by international standards, and that business investment (including FDI) is low by international standards. This chart is from the paper.  In general, what is taxed heavily you get less of.

corp income tax

But this time I was more interested in another of the background papers, this one on the possibility of inflation indexing the tax system.   Even with 2 per cent inflation, failing to take explicit account of inflation in the tax system introduces some material distortions and inefficiencies.  Many of the costs of inflation arise from the interaction with the tax system, and these distortions may be greater in New Zealand than in many other countries because of the way we tax retirement income savings (the TTE system introduced, as a great revenue grab at the time, in the late 1980s).

In the days of high inflation there was some momentum towards doing something about indexation. It had, for example, been a cause championed by former Reserve Bank Governor Ray White.  And in the late 1980s, the then government got as far as publishing a detailed consultative document.  But then inflation fell sharply (and maximum marginal tax rates were cut) and the issue died.  We don’t even have the income tax thresholds indexed for inflation, allowing Ministers of Finance ever few years to present as a tax cut an increase in revenue that should never have occurred in the first place.

In the early days of inflation targeting there might even have been a case for letting the issue die.  The inflation target was centred on 1 per cent annual CPI increases, and that target was premised on a view that the CPI had an annual upward bias of perhaps as much as 0.75 per cent per annum).  But since then, the extent of any biases in the CPI have been reduced, and the inflation target has twice been increased.   The inflation target now involves aiming for “true” inflation” of at least 1.5 per cent per annum.

The distortions are most obvious as regard interest receipts and payments.  Take a short-term term deposit rate of around 3 per cent at present.  Someone on the maximum marginal tax rate (33%) will be taxed so that the after-tax return is only 2 per cent. But if, as the Reserve Bank tells us, inflation expectations are 2 per cent, that means no real after-tax return.  Compensation for inflation isn’t income and it shouldn’t be taxed as such.  Only the real component of the interest rate (1 per cent) should be taxed.   The same distortion arises on the other side, for those able to deduct interest expenses in calculating taxable income: in the presence of inflation, this tax treatment subsidises business borrowing.  The amounts involved are not small.   As economist Andrew Coleman notes in his (as ever) stimulating TWG submission

Even at low inflation rates, these distortions are substantial. In 2017, for instance, residential landlords borrowed $70 billion. Even if the inflation rate is as low as 1 percent, this means residential landlords can deduct $700 million of real principal repayments from their taxable income, a subsidy worth over $200 million per year. New Zealand households lend in excess of $150 billion. When the inflation rate is 1 percent, lenders are expected to pay tax on $1.5 billion more than they ought. Many people who invest in interest-earning securities are elderly, risk averse, or unsophisticated investors. For some reason the New Zealand Government believes these investors should pay more tax than any other class of investors in New Zealand. It is a strange country that taxes the simplest, most easily understood, and the most easily purchased financial security at the highest rates. It suggests the Government has little interest in equity, its protestations notwithstanding.

There are other distortions too, notably around trading stock valuations and asset valuations on which true economic depreciation would be calculated.

As reflected in the paper released this week, officials are very wary about doing anything about fixing these distortions (and they fairly note that “no OECD country currently comprehensively inflation indexes their tax system”), and they devote many pages to outlining the practical challenges they believe would be involved, and the new distortions they believe would arise from partial approaches to indexation.

I have some sympathy with the stance taken by officials on the specific challenges to doing comprehensive indexation, especially in a way that does not bias transactions through favoured institutional vehicles.  But it is a particularly bloodless document that seems to reflect no sense of the injustice involved in taxing so heavily relatively unsophisticated savers (while subsidising business borrowers, especially those financing very long-lived assets).

This seems like a case where some joined-up whole-of-government policy advice would be desirable.  There would be no systematic distortions arising from the interaction between inflation and the tax system if there was no systematic or expected inflation.   Systematic inflation isn’t a natural or inevitable feature of an economic system –  in some ways it is about as odd as changing the length of a metre by 2 per cent a year, or the weight of a gram by 2 per cent a year.  In the UK, for example, (and with lots of annual variation) the price level in 1914 was about the same as it had been in 1860).  And the most compelling reason these days for targeting a positive inflation rate is the effective lower bound on nominal interest rates, itself created by policymakers and legislators.   Take some serious steps to remove that lower bound and (a) we’d be much better positioned whenever the next serious economic downturn happens, and (b) we could, almost at a stroke, eliminate the distortions –  and rank injustices –  that arise from the interaction between continuing, actively targeted, positive inflation, and a tax system that takes no account of this systematic targeted depreciation in the value of money.

It wouldn’t be hard, but our ministers, officials (Treasury and IRD), and central bankers currently seem utterly indifferent to the issue.

7 thoughts on “Inflation and the tax system

  1. Michael, I agree with you 100% that “Systematic inflation isn’t a natural or inevitable feature of an economic system – in some ways it is about as odd as changing the length of a metre by 2 per cent a year, or the weight of a gram by 2 per cent a year.”

    Oh Michael, you’re sounding less like an economist and more like an engineer!

    I’d like to see an inflation target of 0%, and I’d also like to see land prices brought back into the CPI calculations.

    Since the RBNZ was nationalised by the first Labour government in the mid 1930s, the general level of prices has increased by a factor of about 70 times.


    • I’m more inspired by the biblical injunction to maintain honest weights and measures

      The parallel isn’t exact, and I think there is a role for monetary policy in managing recesssions/booms (for which there is no parallel in weights/measures), but as a starting point it is useful.

      I wouldn’t want to include land in the CPI. Doing so – under current RMA rules etc – would be likely to engender excessive and unnecessary volatility in monetary policy. I’d prefer to fix the land issues directly, with a presumptive right to build (rather than councils telling us where houses and can’t be built).

      Liked by 1 person

  2. Inflation targeting through suppression of interest rates is an integral part of monetary policy throughout the world. The official rationale is that consumption and economic activity are increased (buy now rather than later when goods may cost more). In practice the main beneficiaries are governments, banks and the wealthy.. Governments in particular can borrow at concessional interest rates, and debt is constantly devalued by institutionalised inflation ( which they write into their monetary policy). Likewise the owners of property and other assets can borrow below real market rates, and have their debt devalued over time.
    Banks earn their wealth by the process of credit creation and lending, and likewise earn excess profits when interest rates are suppressed by monetary policy. Savers and those who own no assets miss out, and are taxed on wage increases and interest earnings, with no adjustment for inflation.

    Liked by 1 person

  3. Hi Michael, Thanks for your detailed answer to my inflation/deflation question on your ‘A Picture in continuity’ article, reading this article had further increased my understanding and my question was driven from my feeling that ‘Systematic inflation isn’t a natural or inevitable feature of an economic system’. I have a gut feeling that we don’t benefit as much as we should from the deflation of all of the things which have crashed in price from technological gains, the china effect etc.. My suspicion is that inefficiencies in central and local government and increased regulation also soak up a lot of these gains.

    This is really an incredible blog, I’m half way through the US debt default article, very professionally put together. Great article on the statistics department too, it appears the dollar moved a whole 2 minutes before the official release and had finished its move by the time of the release. A case of quaint NZ naivety perhaps however that is not an excuse!

    Perhaps you would have time to comment on this; The difference in yield on a 2 year US treasury bond and a 2 year NZ fixed mortgage rate is currently about 1.5%. This just fundamentally doesn’t seem right? That’s saying the risk premium of lending to a guy with an average job to buy a decrepit property in provincial NZ is 1.5% more than lending to the US government commanding taxation over a 21 trillion economy and the the most powerful military of all time. Again what am I missing! Thanks.

    Liked by 1 person

    • Thanks Rob

      On your question in the final para, there is no obvious natural relationship between the two rates you quote. And the US 2 year bond rate is about 2% more than the Italian one, when a serious crisis (including the explosion of the euro, and/or departure of Italy must be a non-trivial possibility in the next few years).

      Recall that banks are lending with a whole of portfolio view in mind, and residential lending (when the market isn’t skewed by the govt) tends to be pretty low risk for banks, so across the portfolio there isn’t a need for a large risk margin. And, on the other hand, the US fiscal situation is looking increasingly reckless and worrisome: recall that Roosevelt’s default in 1933 wasn’t because the US couldn’t pay, but because it chose not to. And that is so with most sovereign defaults (at least among advanced countries).

      But the other important consideration is that the interest rates you quote are in two quite different currencies, and the basic yield curve in each currency is determined by economic conditions in that particular economic zone (supply of savings vs demand for investment, expected inflation, and risk). Both the US and the euro area have big fiscal problems, but the euro area yield curve is well below that of the US at present because Europe has little population growth, and high savings rates, and not much policy stimulus. In the US, savings rates are low, the population is still growing moderately, and Trump’s fiscal stimulus is providing a (probably temporary) sugar high.

      Having said all that, I wouldn’t regard 0.8% as acceptable compensation for Italian govt risk, whereas 2.8% probably is reasonable compensation for US govt risk.


  4. I don’t think they are indifferent to the issue of tax and inflation rates, it is a deliberate policy. As you say, most people cannot see the effective increases in taxes and see adjustments of thresholds as “tax-cuts”- a perfect system for governments.

    In response, rational actors invest in assets that rise with inflation long-term and avoid term deposits. If they investing in property, income from the asset rises with inflation and the debt is inflated away. If the current home interest rate is 4% with 2% inflation, the real interest rate is 2%. As you say, the return of a term deposit of 3% after-tax 33% (2%) and inflation is close to zero. This distortion of the market for interest rates has led to bubbles in nearly all asset markets.

    This policy results in transfers of wealth from the saver to the borrower. The largest borrower is generally the government.

    “I have some sympathy with the stance taken by officials on the specific challenges to doing comprehensive indexation”. I really do not see the challenges, start by indexing tax thresholds to inflation rates retrospectively annually. Do not tax interest earned and do not allow interest paid to be deducted as expenses.


    • I’d certainly support indexing the thresholds for income tax – in fact, it is such an obvious step it is shameful it hasn’t been done. The document I linked to highlights various practical problems with actually attempting to inflation index components that lead to taxable income itself. In general, I favour a move to a consumption tax (which is what not allowing deductiability/assessability for interest is a step towards), but i – and they – were writing in the context of an income tax system. Fixing the treatment of interest without fixing depreciation just sets up new distortions (because for a business, the two tend to be offsetting – firms can deduct too much interest, but not enough depreciation. And one could do a fix for small savers, but actually the issue isn’t so large for people with a 10% tax rate, relative to how much it matters for those on a 33% rate.

      In theory, with appropriate and symmetrical adjustment all round, taxing nominal interest (and deductability of nominal interest rates) should simply result in higher nominal rates all round. In practice, those conditions aren’t met.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s