Thursday, July 15, 2010
A Katz-Fink Dialogue on Japan's national debt and the management of national assets
The other day I asked two of the sharpest Japan optimists, Naomi Fink and Richard Katz, to give me five sentence-long answers to two simple questions on the actual size of Japan's national debt and the management of its national assets. They both grossly disregarded my instructions as to length, as you will see. I am the wiser, however, for their disobedience. With their permission, I here share their answers to my two questions.
Q1: Why is net debt rather than gross debt the more important number to keep in mind regarding Japan's financial position?
Q2: What are the main assets aside from Japan Post and shares in JT that the Government of Japan should sell in order to finance its budget deficit, staving off for a while the imposition of a rise in the consumption tax?
* * *
Richard Katz, Editor-in-Chief, The Oriental Economist
"The financial burden of the debt equals the net amount that the government as a whole owes to the private sector (and therefore the net amount of interest it has to pay the private sector). But about half of the gross debt in Japan consists of debts that one government agency owes to another, debts that in fact cancel each other out. For example, a substantial portion of the gross debt consists of bonds bought by the Bank of Japan rather than private investors. Another substantial portion of the debt consists of surpluses in the social security account. Suppose the social security system raises 100 billion yen more in year X than it spends. And suppose the rest of the government spends 100 billion yen more than it gathers in taxes. For the government as a whole, the deficit that year is zero. But in Japan, the social security system will "lend" that 100 billion yen to the MOF. The government's gross debt has thereby increased by 100 billion, even though the net debt--and hence its real financial burden--has not increased at all.
Selling assets does not solve the problem, except in a very short-term slipshod accounting sense. Let's assume that the assets are worth something, that they earn a return. Otherwise no private buyer would want to buy them. In that case, the government has gained some immediate cash, but it has lost the right to all the revenues that would accrue to it in the future from owning that asset. So, for the sake of quick buck, it has reduced its future non-tax revenue. There are some very good economic reasons to get the government out of the business of selling cancer sticks and running a huge bank and insurance company. But the notion that solves the problem that annual revenue falls short of annual spending is not one of them."
Naomi Fink, Japan Strategist, Global Marketing & Trading Division, Bank of Tokyo-Mitsubishi UFJ, Ltd.
"Regarding the first question, it is not more important to keep in mind net than gross debt when looking at the adverse impact of a massive government balance sheet upon growth - that the government carries much intra-governmental debt with little to show for it is not a recommendation for the efficiency of their asset allocation. However, if we are to discuss the probability of a fiscal crisis, canceling out intra-governmental assets and liabilities is relevant first and foremost - as Rick Katz pointed out, the MOF's ability to borrow from the social security system reduces the need for fund-raising from the private sector. But apart from net versus gross debt, it is also important to keep in mind when discussing the probability of financial crisis the country's external surplus/deficit position. Much less of Japan's current account surplus is being "recycled" abroad via direct and portfolio investments and instead is coming back to sit in banks as savings. Then, instead of lending banks are buying... what else but government debt? These "risk free" assets gain a higher capital weighting on banks' balance sheets, plus allow absorption of a greater amount of debt. This is why the G-20 exempted Japan from the necessity to reduce its fiscal deficits.
On the second point, I would disagree with Richard that selling assets makes no difference to deficits - in the late 80's when the government privatised NTT , there was not only the $70-80bn gained via the series of offers from privatisation itself (which is a sizable flow over the space of three years) then there was also a rise in both corporate and income taxes - because you can tax a privatised company's profits (assuming that privatisation makes it more efficient and thus profitable) plus the dividends on mostly retail investors' participations in the IPO. Longer-term, the idea is that these flows will jump-start asset reflation, which will be extended by the greater productivity of assets under private-sector than public-sector management. And judging by the experimental reverse privatisation of Yu-pack, such an argument might be made for Japan Post! Apart from Post and JT, one potential target would be the highways (perhaps follow the JR model and privatise plus regulate, to minimise profiteering as the government is wont to do). Then you have several hundred trillion in financial assets, comprised of foreign reserves, stocks (from the Banks Shareholding Purchase Programme), Zaito debt and FILP deposits. Of course, it would not make sense to dump all of this at once, but selective thinning out of the portfolio as assets reflate would make sense - these assets were accumulated under emergency circumstances, after all. I think that was the idea that Fortress had when they proposed setting up a fund that would buy non-performing assets from banks and - yes - governments projecting a 20% return. Yes, they might be wildly optimistic but let's face it, asset valuations are quite low in Japan and there is plenty of room to maximise the productivity of some of the currently government-owned or supported assets instead of using fiscal stimulus to keep zombie firms from going bankrupt.
Richard Katz -
The statutory tax rate on corporate earnings is 40%. The actual average tax:profits ratio for the corporate sector as a whole is closer to 26%. If the government owns NTT, it gets all of the profits. If it sells it and taxes the profits, over the long haul, it gets a fraction of the profits. Unless NTT more than doubles or triples its profitability simply by virtue of not being government-owned, there is net loss in the future revenue stream for the government. Keep in mind there is a big difference between government ownership and government management. Management is JR running too many mostly-empty trains due to government pressure. Government ownership of shares of a firm run like a private one is a very different kettle of fish.
No question but that the government should get totally out of certain businesses because that will raise efficiency and potential GDP growth. But the impact of that in Japan is rather marginal given the low share of government-run businesses. Without a genuine program to raise the efficiency of the private sector and deal with chronic shortfalls in domestic private demand, Japan will remain addicted to deficit. One-time only sales of assets won't solve the problem--no more than does a sale of a division by a private firm that still doesn't know how to make money. It only postpones the inevitable.
As for the claim that the government selling its shares in JT will instigate asset price inflation, I'm sorry, but I don't get it.
I believe Japan Postal Bank and Japan Post Insurance should be abolished, not privatized. Turning one of the world's biggest governmental monopoly banks and insurers into private monopolies doesn't promote efficiency.
As for the other financial assets mentioned (foreign reserves, stocks (from the Banks Shareholding Purchase Programme), Zaito debt and FILP deposits) how does transferring ownership of pure paper promote growth if the real assets behind this paper remains problematic? It seems to me that debt is debt, whether owed by the government or the private sector. We've seen lots of private debt crises, e.g. the US subprime. The issue is whether the assets backing that debt create the financial wherewithal to finance the debt service. If the government borrows money to improve infrastructure that, in turn, raises GDP and tax base, the debt is justified and does not become a financial problem. If it created debt to build bridges to nowhere, it does create a problem. If the private sector creates debt and debt-backed securities on buildings that are worth less than what it cost to build them, then it does create nonperforming loans. Just transferring debt backed by bad assets from government hands to private hands does not create financial solvency.
Naomi Fink -
Rick, I do agree that increase in efficiency is central to the success of privatisation - if it not possible at once to maximise profits and reduce costs by electing private-sector management, privatisation is, as you say, a mere transfer of assets. More transparent and better-quality management of the assets does matter and one clear purpose of privatisation should be to raise the marginal revenue of the privatised entity.
As such, when discussing the merits of offloading the business from the government's balance sheets as a going concern, one must consider both sides of the balance sheet - if the government achieves poor marginal utility from the assets with regard to their pareto efficiency, then the government's costs are also likely to be larger than those which would be imposed upon a more efficient manager of those assets. And I do not take as a foregone conclusion that private-sector management, even if subject to government regulations, will necessarily achieve a negligible change in efficiency of asset allocation.
You say, "No question but that the government should get totally out of certain businesses because that will raise efficiency and potential GDP growth." Indeed - and in the final paragraph, you question the ability of private sector Japan to manage entities, once privatised. Certainly, corporate governance is a very valid concern, shared by many overseas investors in Japanese stock. But low levels of current productivity is precisely why many wish to get in here - to achieve the superior returns that accompany increases in marginal revenue. This is where I believe that widening the investor base - particularly opening the doors to a variety of foreign investors - might help. The Fortress example is one such initiative (though time will tell whether successful). While the government is reluctant to "sell out" to foreign investors, there is probably value in exploring a model wherein foreign investors provide valuable information on the efficiency - and thus pricing - of private sector assets, long skewed by indiscriminate government handouts to for-profit firms.
As for your caveat that the impact of government getting totally out of certain businesses in Japan, that it is rather marginal given the low share of government-run businesses -- sure, but public financial intermediaries DO control 30% of domestic lending, which has the same effect - if we are to believe Modigliani and Miller, firm valuation is not determined by whether the firm is capitalised by debt or equity. The high rate of government intervention in lending could explain at once the high level of tolerance for poor corporate governance as well as the low level of competition among private sector lenders, who have been increasingly disenfranchised from provision of liquidity.
Regarding the link between IPO's and asset reflation - I refer back to NTT and the large proportion of retail investor participation in the offer. Retail investors, large holders of cash, dissave (thus reversing one of the key drivers of asset deflation) and divest deposits into equity assets. Thus, bank deposits do not build and are NOT funneled by banks into JGB holdings (Yucho was the leader of this trend), a trend that continues until investors turn risk averse once again about the ability of their future income flows (investment as well as employment income is important in a country with an aging population). So I take your point - the IPO is no universal salve to poor asset allocation and deflation but is one trigger which, if followed through by greater productivity in assets throughout Japan, could restore a "virtuous circle".
On your point about taxes -- "The statutory tax rate on corporate earnings is 40%. The actual average tax:profits ratio for the corporate sector as a whole is closer to 26%" -- I agree that the tax base is too narrow in Japan - according to the OECD, only one-third of corporates pay taxes at all. But this is a questioning of broadening the tax base, which might be done in tandem with a decrease to the headline corporate tax rate. If a greater number of firms pay taxes of 30% rather than 26% or zero (the remaining two-thirds) then presumably this should minimise the impact of cutting taxes on the handful of large corporates who actually pay 40%.
You say, "I believe Japan Postal Bank and Japan Post Insurance should be abolished, not privatized. Turning one of the world's biggest governmental monopoly banks and insurers into private monopolies doesn't promote efficiency." Perhaps - but the question is whether the government will realistically abolish both - and they probably will not. Thus minimising these entities' inefficiencies is probably a good compromise. There is an alternative to establishing private-sector monopolies and here "privatise and regulate" can work both ways - instead of "running half-empty trains," forcing JP Bank/JP Insurance to adhere to private-sector regulations on disclosure, record-keeping and capitalisation, plus pay private sector taxes could have a positive effect on private sector competitiveness. Using recent examples either actually implemented or proposed, exempting intra-Japan Post transactions from consumption tax puts private sector institutions at a cost disadvantage - as does the universal guarantee on bank deposits, as does the exemption from Basel II (let alone imminent Basel III) capitalization, as does the idea that know-your-counterparty rules should not apply uniformly to JP Bank as to private sector banking institutions. And the argument that imposing these guidelines will merely add to JP Bank's cost base makes the controversial assumption that the benefits of these private-sector guidelines are not valuable contributors to risk-weighted profitability - which if valid must be taken up with the BIS and anti-money laundering task force immediately! Add to this greater disclosure requirements to answer to private-sector shareholders.... greater transparency is central to identifying a firm's operating inefficiencies - such that if JP Bank/JP Insurance truly prove, under private sector standards unable to efficiently deploy their assets, the case for breaking up these large state-controlled "monopolies" emerges much more clearly than it would under limited-disclosure government ownership.
Richard Katz -
Lots of food for thought in what you say. And I completely agree that more foreign direct investment is vital to making the private sector more efficient as it has been in other countries. If I understand you correctly, then you and I are on the same page in saying that a sale of Japan Tobacco or Japan Post Bank simply to gain a quick cash injection does not solve the government debt problem over the long haul. The issue is whether, and by how much, privatization helps raise efficiency in these and other entities
Naomi Fink -
I think we do agree that the quality of asset allocation matters more than the transfer itself of assets from the public to private sector. For example, the idea of privatising half of Japan Post under the Kamei plan most likely would have been, as you say, little more than a transfer of assets veiling a privately-funded expansion of government-controlled monopoly, and it is probably quite helpful in this regard that the USTR and European Trade Commission decided to push for a "level playing field" in postal privatisation.
Richard Katz -
Michael, I think we are done and have given your readers something to think about.
Naomi Fink -
Thank you both very much for this - it has been a good, thought-provoking discussion.