Maestro myth revealed: Alas, as is now widely recognised (also by many central bankers), these actions sowed the seeds for an asset price and credit bubble of gigantic proportions. Since the bubble burst, we have been moving from crisis to crisis - a credit and banking crisis, the Great Recession, an emerging market crisis and now a European sovereign debt crisis. With central banks having long been forced (by circumstances rather than government decree) to open the floodgates and apply an unprecedented series of unconventional measures, it looks more and more likely to us that the confidence crisis in banks and sovereigns will ultimately morph into a crisis of confidence in the central banks who act as lenders of last resort to both banks and governments. Yes, central banks can extend unlimited amounts of credit to banks and governments. But they do so by issuing ever more of their own liabilities - money. And just as the trust in banks' and governments' liabilities eroded when they issued ever more, we believe that the trust in money will erode if central banks issue ever more of it.
Gold and exchange rates illustrate confidence loss: A loss of confidence in the value of fiat money usually shows up first in the price of gold and in exchange rates. By these standards, the process has long started. The price of gold has more than doubled from its pre-crisis levels at the start of 2007, no matter whether it is measured in US dollars, euros or sterling. And exchange rates - the relative price of fiat currencies - have also reflected (relative) confidence losses. Last year, when the Fed embarked on active quantitative easing with massive asset purchases, while the ECB was less aggressive and engaged in the less risky versions of ‘passive' quantitative easing, the dollar weakened significantly against the euro. This year, when the Fed ended its asset purchase programme and the ECB was recently forced to become more aggressive by extending liquidity support to the banks again and starting to buy government bonds (though at a much smaller scale than the Fed) outright, the euro was punished relative to the dollar.
Inflation still low... Meanwhile, the internal value of money, as measured by its purchasing power in terms of goods and services, has been reasonably stable as inflation has remained low, especially in the US and Europe. But this is hardly surprising, given that the global economy only started to emerge from its deepest post-war recession at around this time last year. Moreover, it provides little comfort for the future as global monetary policies remain extremely accommodative despite the economic recovery and as we believe that the sovereign debt crisis will keep both the ECB and the Fed on hold for longer and will thus make it more difficult for many emerging market central banks to tighten monetary policy (see "XXL Liquidity", The Global Monetary Analyst, May 12, 2010).
...but the UK may be a leading indicator: Interestingly, the country with the highest current inflation rate in the G10 is the UK, with CPI inflation at 3.7% by far exceeding the 2% target. Even excluding the VAT hike (but only assuming partial pass-through), our UK economists think that CPI inflation would be running at slightly above 3% currently. Judged by the expansion of the Bank of England's balance sheet since the start of the crisis, the UK has also had the most aggressive monetary policy response, and sterling has weakened significantly versus both the dollar and the euro over the past couple of years. And, as an aside, the UK experience flies in the face of the popular theory that, in order to create inflation, you must have solid credit creation and broad money growth - both M4 and M4 lending (on the BoE's preferred measure that excludes intermediate OFCs) have only barely picked up fairly recently. Recall that similar arguments about preconditions for an economic recovery were put to rest last year when a surge in excess liquidity provided enough traction for the start of the economic recovery. Credit creation and broad money growth or the absence of economic slack would certainly aid and abet inflation, but their absence does not necessarily rule out inflation and inflationary risks when such expansionary monetary policy regimes are in place.
ECB the latest victim: More recently, the focus has shifted to the ECB, which has traditionally been viewed as the most independent and credible among the major central banks by many market participants. To a large extent, the ECB's credibility has been built on its very nature of a supranational central bank mandated with the primary goal of price stability, issuing a denationalised currency remote from government influence. However, the credit crisis that has morphed into a sovereign crisis has forced the ECB into actions that threaten to undermine its credibility over time.
Credibility undermined: First, the ECB had to make a climb-down on its collateral rules by accepting Greek government bonds irrespective of their credit rating. While justified as an exceptional step, it seems clear that the ECB will not be able to deny similar treatment to any other member state getting into trouble. Second, the decision to buy government bonds in the ‘dysfunctional' secondary market was a big step, as can be seen by the open controversy it created in the ECB Council. While these purchases do not violate the Maastricht Treaty, which only rules out direct lending to governments or bond purchases at auction, they clearly help governments finance their deficits at lower rates than otherwise. Third, by continuing to provide banks with unlimited liquidity at various maturities, the ECB keeps even the weakest players afloat and thus slows down the necessary consolidation and recapitalisation in the banking sector.
Hostage to financial and fiscal stability concerns: Of course, all these measures can be justified with concerns about domino effects and thus a potential meltdown of the system, and it is difficult to see how the ECB could have acted differently in these situations. Yet, these actions also delay the necessary adjustment as banks and governments have learned to rely on the ECB as a lender of last resort. And without the necessary adjustment in the form of banking sector consolidation or the threat of regulatory intervention (breakups, orderly wind-downs, etc.), recapitalisation of banks (especially those in public ownership) and major fiscal reforms, the ECB will likely remain hostage to financial stability concerns and find it difficult to respond to inflationary pressures when they appear.
Vigilantes asleep: In short, over the next several years, central banks will face an uphill battle to defend their credibility, and many of them will have their hands tied by financial and fiscal stability concerns. Amazingly, though, while the gold price and currency markets have been sensitive to these concerns, bond markets appear to be lulled in by liquidity, carry and roll-down that seems to springs eternal. But this is not new: bond markets also took years to take on board the Great Inflation of the 1970s and the big disinflation of the 1980s and 1990s. The famed bond vigilantes are fast asleep, again.
Metabolism: lower now, higher later. PM Hatoyama's resignation will both slow and speed policy metabolism. The resignation will slow metabolism in the sense that it will be very difficult to get bills through the Diet until the DPJ's leadership issues are settled. Choosing a new party leader could be difficult. The resignation may speed metabolism in the sense that it has ended debate on whether he will resign, and thus allows political evolution to move to the next stage more quickly. Historically, increases in political metabolism have been good for Japanese assets.
As for potential successors, it seems unlikely that many senior DPJ leaders would want the top job now, since the new party leader could be held responsible for the likely defeat of the DPJ in the Upper House election in July. Thus, a less well-known candidate may emerge, as an interim solution. That said, a senior leader with a long history in the party, such as Finance Minister Kan, could also be drafted. So long as Mr. Ozawa, now Secretary General of the DPJ, remains in a position of power in the party, whether official or otherwise, the basic stance of the DPJ is unlikely to change, regardless of who becomes PM. Thus, Japanese press reports are focusing on leadership changes in the DPJ.
Immediate policy implications:
a. Pressure on the BoJ for further easing will likely evaporate, until politics settle. Hence, yen weakening becomes less likely over the summer.
b. Key pieces of legislation may not pass, and would have to start from scratch in the next Diet session. At the moment, two key bills are under debate - the postal reform reversal bills and the labor law revisions to prohibit use of temporary workers in manufacturing. Since both bills are not business-friendly, their failure to pass the Diet may be regarded as positive news in financial markets.
c. Timing of selection of the new DPJ leadership is hard to foresee. One reasonable scenario is that it might take a week for the DPJ to choose a new leader. Another is that the DPJ Diet members may meet as early as this Friday, and choose the new DPJ President - with no role for regional party members. In this scenario, the Diet would convene sessions quickly to appoint the new DPJ leader as PM. The new PM would likely retain most current Cabinet members, in order to return quickly to focus on the upcoming election in July.
The timing of DPJ leader selection will impact whether there is a Diet session extension. With a Diet extension, the DPJ could decide to postpone the election until either July 25 or even August 8. Such extensions would likely be adverse for financial markets, because they would revive the possibility of passing market-unfriendly legislation. If the DPJ does not extend the Diet session, passing the postal and labor bills could be difficult. If the DPJ does extend the Diet session, it would be easier to pass these bills. The relationship of the new DPJ leadership and coalition partner People's New Party will be crucial in these decisions. Until the new DPJ leadership is chosen, it is hard to predict the Diet schedule or the election schedule.
d. The longer-term direction of economic policy depends on the outcome of the Upper House election, which in turn depends on what happens to the leadership of the DPJ. We think that the most likely outcome of the election will be a coalition of the DPJ, the Komeito and the new party formed by former Welfare Minister Masuzoe. Such a coalition would be unlikely to put strong pressure on the BoJ or make significant changes in other policies. It is possible that leadership changes in the DPJ might allow a political reconciliation between the DPJ and the newly formed Your Party under former Government Reform Minister Yoshimi Watanabe. Such a coalition would likely put strong pressure on the BoJ and could change the direction of both fiscal and structural policy to be more market-friendly.
Introduction and Summary
Japan's Upper House election in July will be crucial in setting policy expectations among investors over the next few years. There is huge uncertainty at the moment, however, because the potential outcomes imply a policy barbell: The heavier end of the barbell implies policy stagnation. The lighter end of the barbell implies much increased pro-growth policy metabolism. This conclusion comes from the following expectations.
Analysis starts with the current coalition, now reduced to two parties in the wake of the departure of the Social Democratic Party (SDP) in a dispute over military policy. The senior coalition partner, the Democratic Party of Japan (DPJ), is likely to win only 93-107 seats compared to 116 now. The junior partner, the Kokumin Shinto, is likely to end up with only 3, compared to 6 now. Such an outcome would leave the coalition far short of the 122 seats needed for an outright majority in the Upper House.
Such an outcome means that passing legislation under the current coalition would be very difficult. With SDP out of the coalition, the total strength of the coalition in the Lower House falls below the two-thirds needed to override bill rejection by the Upper House (provision in Article 59 of the Constitution). Therefore, passing legislation in the Upper House is essential, but would require case-by-case coalition building - a very time-consuming process.
Of course, the coalition might change. However, the most likely outcome of a coalition change would involve a DPJ coalition with the Komeito and the Kaikaku parties (DKK coalition). Our reading is that such a coalition would not raise pressure on the BoJ significantly, and would not likely adopt aggressive fiscal reform or productivity enhancement policies.
The other end of the policy barbell would come if the political hurdles to a coalition of the DPJ with Your Party (YP) are overcome. A coalition of the DPJ with YP might trigger a major swing of policy towards market-oriented, productivity-enhancing policies, with more pressure on the Bank of Japan for aggressive monetary easing.
We derive these conclusions from a top-down model of the election. The methodology, the base case results and scenarios are discussed below, in turn.
Methodology
Our model is built on some key elements of the current Japanese political scene:
a) There are two large parties, and a myriad of small parties. The larger parties tend to take a somewhat larger share of seats than of votes, in light of the seat allocation rules.
b) Some voters have strong party loyalty, in light of interest group connections, but a large and growing share are ‘floating voters', who are willing to cross party lines.
c) The electoral system is split into a proportional constituency where seats are allocated to parties on the basis of their percentage of the overall vote, and a set of election districts (based on prefectural lines), with two or more seats depending on the population of the prefecture.
d) The parties have clear differences of policy stance, and have histories that make cooperation difficult for some and easy for others.
In light of these characteristics, the first step is to correlate the share of each large party in public opinion polls to the actual share of seats received in the election. This is done with a statistical technique that correlates public opinion poll support rates into shares of seats, accounting for the economies (or diseconomies) of scale. For the proportional seats, the equations imply that the sum of DPJ and LDP seat shares will be 50%, and that the LDP will garner only 39% of these. For the election district seats, the results imply that the sum of the DPJ and LDP seats will be 54%, and that the LDP will take 44% of these.
Two technical points are of importance. First, our estimations use public opinion poll data from one month prior to the election. Because the election will be in July, but we only have May data for the opinion polls, we have estimated support rates for June, based on extrapolation of recent trends. Second, in making our forecasts, we mixed both Upper House and Lower House elections, in order to increase the degrees of freedom in the regressions. The cost of this is that there may be an element of mixing different types of elections into a single pot.
The next issue is how to determine the seats for the small parties. Elections with so many small parties are quite rare in Japan. However, the 1993 Lower House election was similar, and we used the results as the basis for the current forecasts. The shares of seats for the smaller parties in the 1993 election were quite similar to their support rates in opinion polls. Hence, we have taken opinion poll support rates for the small parties, extrapolated recent trends, and used these to predict seat shares for the small parties.
With the seat shares for both the large and small parties thus calculated, it is a matter of arithmetic to derive seat numbers for the parties. For the proportional district, we use the vote shares as the basis for d'Hondt system calculations of seats. For the election district seats, we do the same, with an important proviso: We assume that each party will name enough candidates to fill whatever seats it wins in each district.
Once the numbers of seats are determined, the next step is to see what coalitions are possible to reach a majority of seats. The formation of coalitions has two aspects, the numbers and the philosophy. It is necessary for both aspects to work in order to form a viable coalition.
The number aspect is easy: It is only necessary to calculate whether a given combination has a majority of seats. The Upper House has 242 seats, and thus it is necessary to have 122 (i.e., one more than 50%) to achieve a majority. For a stable majority - i.e., one large enough that neither individuals nor groups have a credible threat of bolting and thus destroying the majority - requires more. There is no exact figure that defines a stable majority, but a figure above 135 would generally be regarded as sufficient.
The philosophy aspect of coalitions is harder. This aspect involves not only stances on policy issues but also the approach to leadership, the degree of consultation needed to maintain cohesion, the history of the parties, and the personal chemistry of the leaders and party members.
Therefore, determining possible coalitions involves overlaying two matrices, one with the numerical compatibility and another with the philosophical. Only when both aspects work is a coalition possible. In our base case and scenarios below, we use different numerical outcomes and different philosophical matrices to search for viable coalitions.
Election Results: Base Case
We applied the method described above. The results show that the DPJ loses 17 seats, and the LDP loses 10. The small parties gain 27 seats, net. The biggest gain comes to Your Party (22), and the next largest to Kaikaku (9).
These numbers have several significant implications. First, the current coalition will not likely have a majority in the Upper House. Hence, in order to pass legislation, it would have to rely on a two-third majority in the Lower House; however, now that the Social Democrats have left the coalition, the DPJ coalition no longer has a two-third majority in the Lower House. In short, these results imply legislative stagnation, should the current coalition remain intact.
The second implication is that a new coalition might become attractive to the DPJ. It is therefore necessary to hypothesize about what coalitions are possible, and what each might mean for policy.
In our base case, the DPJ gets 99 seats, and thus the only two-party coalition that could form a majority in the Upper House is DPJ-YP, and then only barely. Among the three-party coalitions, a DPJ alliance with the Komeito would create potential majorities with several other single parties, as would a coalition with YP.
Numbers matter, but so does philosophy. The DPJ could form coalitions with Komeito, the Kokumin Shinto, Kaikaku, and Tachiagare parties without major philosophical or political problems. However, coalitions between the Communists and Social Democrats would be hard, in light of differences over defense policy. In addition, a coalition with YP would be difficult because of political factors and differences on fiscal and structural reform policies.
Overlaying the arithmetic and philosophy matrices narrows the potential post-election coalition alternatives. There is only one, DPJ+Komeito+Kaikaku. Assuming that the philosophical differences of the YP and the DPJ are not overcome, this is the only workable coalition.
Given the philosophical stances and the political styles of the three parties in such a coalition, it seems unlikely that major changes of policy would occur. On deflation, such a coalition seems unlikely to raise pressure on the BoJ, in light of the personnel in the coalition partners and political histories of the leaders. Nor would there likely be major changes in the orientation of fiscal reform policies or productivity/deregulation policies.
Scenario Analysis
The results above are based on our point estimates of the election outcome. However, the point estimates are subject to error for several reasons. First, the public opinion poll data could be in error, in light of sampling problems or reluctance of respondents to give frank answers; or, our extrapolations could be in error. Second, the sample of elections that we used goes back only 20 years; the number of degrees of freedom is small. Third, the structure of the model may be inadequate. In light of these potential errors, we have calculated scenarios based on two standard-deviation bands around the initial seat forecasts for the major parties. We do bull and bear scenarios (defined as bullish or bearish with respect to the outcome for the DPJ).
In general, the worse the DPJ does, the better the outcome for YP and Kaikaku, and vice versa.
These results only give the arithmetic, however. As previously, the arithmetic must be combined with philosophical consistency in order to get an overall picture. Using the same philosophy matrix as before, but substituting the DPJ bull/bear scenario numbers, the following possible coalitions exist for the bull and bear scenarios.
In the bull case, the DPJ could form a two-party coalition with the Komei party, although the majority would be very small. With the Komeito, addition of the Kaikaku would give a large majority in the Upper House. In the DPJ bear case, there are no two-party coalitions that bring a majority. The DPJ would have to bring both the Komeito and the Kaikaku parties into a coalition. No other three-party coalition is both numerically and philosophically possible.
A particularly interesting scenario emerges from changing the philosophy matrix. At the moment, the philosophical and political compatibility of the DPJ and the YP is low. However, should a large defeat for the DPJ bring leadership changes in the DPJ, the two parties could become compatible. We illustrate the potential coalitions in this case. The numbers of seats are the same as in the other DPJ base case, but the compatibility matrix is changed to allow coalitions of the DPJ and YP and others.
Read Full Article »