Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 29 July 2013

Japan and the consumption tax

While most international attention has focused on recent developments in Japanese monetary policy, there are interesting developments on the fiscal side too. A key issue is the proposal to raise the national consumption/sales tax from 5% to 10% in two stages beginning in April next year. Japanese Prime Minister Shinzo Abe says he will wait until probably the autumn to make a final decision, and the macroeconomic outlook will be a key factor. The proposal has the support of Bank of Japan governor Haruhiko Kuroda. However the more interesting question for Kuroda is how the Bank will react to the sales tax increase.

Much of the reporting on this issue is along the familiar lines of whether it is better to focus on reducing the government’s very high level of debt (raise sales taxes) or ending deflation in Japan (don’t raise sales taxes). While this debate is a familiar one, there is an additional twist with a sales tax. An anticipated increase in sales taxes, by raising expected inflation, will - other things being equal - provide an incentive for consumers to bring forward their spending. Macroeconomists would describe this as a real interest rate effect, but in simpler terms it makes sense to buy before prices go up.

This incentive effect has been observed in Japan in the past, and in other countries. (See page 12 of this IMF report on the issue.) The UK cut VAT for just one year in response to the recession in 2009, a measure I have described as New Keynesian countercyclical fiscal policy, and this may have raised consumption by over 1%, in part because consumers anticipated that prices would rise again in 2010. (The over 1% figure comes from here, although this analysis is more conservative.)  

However, this effect only occurs if monetary policy does not react to the sales tax rise, and the increase in headline inflation that this will bring. If every percentage point increase in inflation is matched by the same increase in the nominal interest rate (and we ignore taxes), the effect will disappear. (Prices will rise, but so will the value of my savings so I can afford to wait.) If the Bank of Japan attempts to reverse the increase in inflation by tightening policy still further, then we get a very undesirable outcome.

These considerations suggest two things. First, if they take place, increases in sales taxes should be deferred by long enough to allow any bringing forward of spending to happen. The worst thing you can do in current circumstances is implement an unexpected sales tax hike. Why not raise sales taxes by 1% each year for the next five years? Those who suggest that acting gradually ‘risks losing the credibility of financial markets’ should be ignored. Second, the Bank of Japan should commit to ‘see through’ the impact of sales taxes on inflation, and not tighten monetary policy in any respect (conventional or unconventional) following the increase in inflation that higher sales taxes will bring. It would be good if that commitment can be made publicly before the increase in sales taxes is confirmed.



  1. Japan has let its debt rise too much so it is hard to see how this thing will work. Taking reasonable estimates for when things work and when they don't the country will drive against a wall. Only not clear which one or a bit from both.
    -2% inflation, 2% real growth, interest 4%, debtlevel at present 240%. Interest is the major one of course. Hard to see how they can keep the rate longer term below 2+2, with a Japanese economy working again and likely the rest of the world as well except possibly EZ. You not only have to buy up new issues but when the economy is running and even before that as we see now, also 'old' bondholders looking for yield.
    Japan is likely the after the EZ worst patient.
    Anyway you end up with 9.6% GDP interest (240x4%). While you inflate away and grow away resp 4.8% (2.40x2% plus 2% (2%of100%) equals 6.8%. 2.8% short, so debtlevel will rise 2.8% just because of the interest.
    Plus a government that is 10% GDP minus now and is planning a stimulus on top of that. So likely higher than 2.8%.
    -Doesnot work. Debtlevel will rise even more because of lower growth and/or lower inflation. Plus the cost of the then failed stimulus on top of that.

    Basically even with 5% salestax debt will rise. Stimulus simply looked to be financed by tax rise (and lower consumption)anyway.
    Plus on top of that stimuli in Japan have been done for decades now and are extremely inefficient.

    Basically Abe is elected because the standard oppositionparty who was as an exception in government made an even bigger mess of it than Aba & Co before. Doubtful if this is not successful and if the 5% hits in he will win a next one. Well he is likley gone anyway when we come there.

    What it shows is that you should never get to that debt level in a hardly growing economy with basically by nature no inflation. You either do a General Custer saving attempt earlier when you still have a chance or prepare for a default (with another name). Lesson for all the Southeners in the EZ they are like Japan now too far gone and when not saved by Germany (diorect or via ECB) they will default. With much worse consequences than a huge restructuring.

    Japan looks very French to me, simply waiting till the first group get scared and take a run. When and not if as it looks.
    Subsequently you are in the middle of a huge financial crisis with no easy ways out. Japan inflate with much higher percentages the pensions and savings away of a highly aging population and France has its EZ limitations.

    Back to inflation and VAT. The whole rescue is as said likely to end in tears. I would not be surprised that it will be hard or even impossible to see how this particular part of the plan will work.

  2. Sober View pointed out this weekend that although most countries' workers reach maximum wage just before retirement, Japanese workers reach maximum wage around age 50, after which average wages begin to decline.

    If that's correct, then raising a consumption tax will be especially harsh on workers in the last ten years before they retire. It could make it very difficult for them to add to their retirement savings during their last ten years of working life.

    By the way, the captcha for this posting was 'nutsfu', which I found amusing.

  3. Krugman blog November 17, 2010 'Why I’m Soft On Sales Taxes' says that "people are right, sales taxes are regressive taxes..[but]...we know that countries with strong social safety nets generally rely a lot on consumption taxes."

    Maybe Japan will be a (slightly) fairer country as well as trying to pull itself out of inflation.

  4. An observation I haven't often seen made is that a sales tax, if refundable on exports like VAT, is in some senses a disguised devaluation.

    If you imagine an economy with 20% VAT, which is refunded on exports and charged on imports, the result - other things being equal - is a 20% subsidy on exports and charge on imports. Or like a 20 devaluation without the adverse wealth effect on travel, terms of trade and so on.

    Or am I missing something significant?

  5. @ Rik,

    Don't forget to adjust for the monetization of the debt. The more the BoJ buys back of JGB debt, the better. They simply end up paying interest back to themselves since the State owns the BoJ.

    Monetisation will work wonders for Japan. Do the adjustment and see for Yourself.

    Same for the UK and the US, by the way if they ever get into the same kind of trouble.

    For the EZ it's a different matter since they lack their own central bank didn't understand how important it is to have one (when they constructed the ECB). Now it's too late, it seems.

    1. You are partly right of course.
      However with basically a 10% deficit now and likely to increase with another 5% as stimulus. While sales tax up might be say 1-2% of GDP hard to see how they will close the debt increase. And will fall of the cliff somewhere.
      Another point is that markets are very likely to dump everything Japanese (may be except intenational cies) if it happens at a big scale. All credibility of the CB will be gone. It is simply like high inflation likely just a few percent better than a default.
      Imho Japan is simply too far gone.
      Glad you didnot see the calculation 'mistake', wanted to try if it was not too obvious.

  6. How long does that inflation expectations thing work if inflation doesn't come? Economists in the US have been predicting, and presumably expecting, inflation for almost five years now. The best they can do is lie about the cost of milk. How naive are consumers expected to be, especially if they don't have access to the requisite money for buying things now as opposed to later, for example, when they have a full time job that pays more than minimum wage.

  7. Dear Professor Wren-Lewis,

    I have to agree to your blog entitled by Japan and the consumption tax. (I have been a regular visitor of your blogs for about a year.)

    The Japanese economy still faces a large output gap, no less than 5% of the GDP. The level of the nominal GDP has just started recovering from the bottom of the current business cycle since the final quarter of 2012, thanks to the already widely established three arrows of Abenomics.

    Let me contribute to your blog by suggesting an additional point that the new fourth arrow, which is the proposed sales tax hike to 8% in the name of fiscal consolidation, would be highly likely to damage the recovery. As you may know, the fiscal contraction plan would reduce the nominal GDP by about 1.8% due to the adversely affected private consumption.

    As Keynes put it, boom rather than slump is the best time of fiscal tightening. I regret to admit, however, that Japan never learn the true lessons from world economic history, and is likely to make the same mistake as in 1997, when sales tax was raised to 5%, which triggered the vicious cycle between the financial crisis in Japan and the Asian currency crisis.

    I believe that 2% inflation targeting policy is not a panacea for a miracle sustained recovery for Japan. Rather , I am suggesting a 4-5% nominal GDP targeting policy both though monetary and fiscal measures.

    I am afraid we will continue to go through the infamous stop and go policy, which is unlikely to bring the economy back to the long term sustainable growth path.

    G7 (and G20) may urgently need to come up with true international policy coordination a long the line of Swan Diagram. In fact, I had personally suggested via e-mail this kind of policy ideas to Professor Koichi Hamada, the top economic advisor to Premier Abe.

    Sincerely yours,

    Tomo Nakamaru

    Chief Economist
    Macro Investment Research Inc.
    Tokyo Japan

  8. Let me add to my previous comment the followings:

    Finally, let me express my sincere appreciation for your insights expressed in your blog for Japan.

    I am looking forward to receiving and learning from your wisdom for Japan in particular and the world in general.

    Best regards,

    Tomo Nakamaru


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.