Alcohol deregulation in Japan: Vending machines and FTA-compliant taxes

[Originally posted, with full hyperlinks and various Comments, at]
Japan appeared finally to have recovered from its own financial crisis a decade ago, albeit at the cost of much accumulated government debt. But now the country has been hit by the collapse of its export markets and the rapid rise of the yen, following recession in the US and increasingly world-wide. Professor Iwao Nakatani, former Chairman of Sony, urges a radical shift in economic policy in Japan and elsewhere – from policy ‘based on neo-conservative economics and the philosophy of small government to one based on Keynesianism and welfare state ideology’.
Some may remain sceptical that Japan ever really embraced the former philosophy, and its ascendancy was certainly never as pronounced as in the US, the UK or then Australia – where market liberalisation tended to be linked closely to race politics (Mark Davis, The Land of Plenty, MUP 2008). But deregulation of alcohol distribution is one of Japan’s many transformations over the last decade. It is also the flipside of ever-stricter rules on drink driving, although they also reflect a broader trend towards criminalisation of socio-economic deviance (evident eg in product safety or consumer credit re-regulation).
On the other hand, deregulation is most notable in terms of where you can buy alcohol to celebrate this New Year of the Ox, namely vending machines and those ubiquitous convenience stores. It is less notable in what you pay especially for certain beer-substitutes, which mainly reflects differential tax rates. In fact, such rates may well violate WTO law. Yet there is probably not enough financial reward for potential beer exporters to encourage their home governments to sue Japan. So an implication may be for FTA negotiators (even those now from Australia) to seek some offsetting advantage in their overall bilateral deal with Japan. Yet that would further undermine the entire multilateral WTO framework.

Rising numbers of visitors to Japan (until last September) and other commentators have long remarked on its proliferation of automatic vending machines, including those selling alcohol. Careful observers may have noticed an ID card ‘reader’ supplied with many machines since 2001. Designed mainly to check the age given on drivers’ licences, the readers were partly a response to stricter punishments introduced for liquor store owners selling alcohol to minors. But such machines were also aimed at clawing back market share for ‘mom and pop’ (or: oba-chan and oji-chan) stores. Their share had dropped from 76% in 1983 to 27% by 2000. The big winners had been larger discount outlets and especially convenience stores, after liberalisation of licensing rules particularly in 1993, 1998 and 2003.
But by last year, at least in Kyoto, remaining stores or those operating these machines often seemed to have rendered the readers inoperable (see my photo here). One store owner just told me that they led to too much drop in sales! The local police don’t seem too concerned, now that alcohol is available in so many convenience stores 24/7, although things are reportedly different in parts of Osaka where teenage drinking remains more of a social problem (shakai mondai).
This seems a victory for deregulation proponents, over the opposition of many LDP parliamentarians and their small business constituents, although some health and consumer interest advocates are also now concerned. But this industry turns out to be more complicated.
Of Japan’s large alcohol market, sake (rice wine), shochu (distilled spirits), and dai-san biiru (‘third-category’ beer) each make up about 10%, followed by around 20% for happoshu and 40% for (real) beer. The latter must have malt content of at least 67%, but is taxed the most. In 1994, the year Japan acceded to the WTO, Suntory began marketing beer-like happoshu with malt content of 65%, while Sapporo developed happoshu containing less than 25% malt. Each attracted the following lower tax rates, and hence could be sold much more cheaply than real beer:
Liquor taxes on alcoholic malt beverages (yen per kilolitre)
Malt content Before 1996 Since 1996 {2003-}[2006- ]
Beer: 67% or more 222,000 222,000 [220,000]
Happoshu: 50% – 67% 152,700 222,000 [220,000]
Between 25% to 50% 152,700 152,700 {178,125}
Less than 25% 83,300 105,000 {134,250}
From 1996, however, the government responded by hiking the tax rates for both types of happoshu. In 2003, it also raised tax on happoshu with 25-50% malt content. However, its tax (178,125 yen/kilolitre) and that of happoshu with less than 25% malt (134,250) remained less than that on high-malt happoshu or real beer (222,000, slightly reduced to 220,000 yen/kilolitre from 2006). In 2004, Sapporo (with ‘Draft One’) and Suntory (with ‘Super Blue’) responded with a zero-malt dai-san biiru (‘third-category’ beer), which had even lower tax (and hence retail price) than any happoshu.
In 2002, the then Director of Research at RIETI (a METI offshoot), argued that Japan’s existent tax differentials between happoshu and real beer amounted to tax discrimination between ‘like products’, contrary to WTO rules:

‘Article III:2 of the General Agreement on Tariffs and Trade says: “The products of the territory of any Member imported into the territory of any other Member shall not be subject, directly or indirectly, to internal taxes or any other internal charges of any kind in excess of those applied, directly or indirectly, to like domestic products.” There is no question that beer and happoshu are like products. To the extent that imported beer is subject to tax in excess of those applied to domestically produced happoshu, it is inconsistent with the WTO rules. … This is a lesson the Japanese tax authorities bitterly learned in the 1997 WTO case of Japan – Taxes on Alcoholic Beverages, in which the panel ruled that tax discrimination between shochu and vodka was illegal.
The tax authorities are right in their move to equalize the level of taxation between beer and happoshu. The major breweries opposed to this are wrong. Their argument that happoshu is a totally new product might have been more persuasive if there had been a net increase in the combined market of beer and happoshu, but in reality, happoshu merely substituted some of the beer market. Beer and happoshu should be taxed equally. Of course, the tax authorities could decide to lower the tax rate on beer to the level equal to happoshu, but such a decision is extremely unlikely in view of the current budget crisis.’

Happoshu tax rates were indeed raised from 2003, but some differentials remain. Other commentary and cases in the WTO (against Korean soju, 1997-9; and Chilean pisco, 1998-2000) suggest that key tests for discrimination among ‘like products’ include their physical characteristics, common end-uses, tariff classifications, and ‘the marketplace’ (including possibly evidence from changes in other countries). So Professor Ichiro Araki’s argument in 2002 still seems valid for at least some types of happoshu – and possibly even now dai-san biiru. And that would suggest some persistent limits to liberalising parts of the alcohol industry in Japan, especially when actual or potential foreign imports are involved.
Yet which country and their exporters are really likely to sue Japan? Possibly, low-cost producers of reasonable (real) beer from nearby countries, such as China (Tsingtao Beer), Thailand (Singha), Singapore (Tiger) – and perhaps even Mexico (Corona) or Australia. But China didn’t even join the WTO until 2001, and Japanese consumers went off Chinese food imports especially last year. Singapore, then Thailand and Mexico, now Australia, wouldn’t want to jeopardise FTAs with Japan by launching a WTO complaint over an issue like this. And the situation is further complicated by Japanese brewers investing overseas, especially Kirin (into brewer Lion Nathan) and more recently Asahi (into Cadbury Schweppes in Australia).
So perhaps all that might be achieved against Japan, especially by countries like Australia still negotiating an FTA, is raising the issue to achieve some sort of extra advantage in the overall bilateral deal. But any advantage could be small, given the practicalities of (not) suing just outlined, as well as some questions still about applying the substantive legal test in the context of a product like beer rather than spirits. And such bilateral negotiations anyway would undermine a transparent multilateral system of international trade law, increasingly under threat.
REACTIONS: Ben Bland, “Legal Trouble Brewing for Japan’s Low-Tax Beer”, The Asia File /

Author: Luke Nottage

Prof Luke Nottage (BCA, LLB, PhD VUW, LLM LLD Kyoto) is founding co-director of the Australian Network for Japanese Law (ANJeL), Associate Director (Japan) of the Centre for Asian and Pacific Law at the University of Sydney (CAPLUS), and Professor of Comparative and Transnational Business Law at Sydney Law School. He specialises in international dispute resolution, foreign investment law, contract and consumer (product safety) law.