Q&A on Japan
Today we got the release of Q4 2007 GDP in Japan (detailed break-up here) and it basically toppled all expectations with a whopping 3.7% increase in real GDP on a y-o-y basis (0.9% q-o-q). I remain a bit sceptical of these figures and look forward to the final estimate come the 12th of March. Having said that these figures are not at all in line with what I have been writing on Japan recently. Edward also takes a close look at the Q4 release over at Japan Economy Watch. In that context maybe or perhaps just because of coincidence Francois Guillaume posts a comment to my recent review and preview piece on Japan (version posted at Japan Economy Watch). Since I think the questions as well as the general comments and opinions they convey are very interesting I have chosen to answer them above the fold in traditional Q&A style. Enjoy!
1) GDP much higher than expected...how do you explain that ??
I want to focus on two issues here. First of all, the current figures are preliminary and in this light I expect to see a downward revision in March. In this light, we need to realize I think that given the incoming stream of data we have seen from the three months of Q4 2007 this figure of 3.7% (Q4 YoY) is pretty hard to justify. For a reasonable take on the whole situation Graham Davis from the Economist Intelligence Unit had a good overview I think in his interview with Bloomberg (can’t hyperlink to the clip I am afraid). Also, we should note I think a comment made recently by Takehiro Sato over at Morgan Stanley’s GEF …
Incidentally, in Japan’s case, quarterly GDP data are too volatile to be a suitable criterion for calling the economic cycle. This is clear from the GDP trend in past recessions. Yet while GDP has at times been positive when the economy is in retreat, industrial production has consistently mirrored the downward path of the economy. It seems reasonable to say that the critical factor for assessing the economic cycle is simply the direction of industrial production.
However, if we accept the figure as it is I still don’t think that the underlying path of the Japanese economy has changed even if the level seems somewhat too high. Let me consequently highlight some of the snippets from Bloomberg’s report on the break-up of the GDP components as well as the much more detailed break-up provided by Edward vis-à-vis the official estimate provided by the Cabinet office. What thus seems clear to me is that although consumption rose in the last half of 2007 net exports and by derivative capex continues to drive forward Japan on the margin. Remember that we need to talk about levels here too since private consumption commands a much larger share in the Japanese economy than does both investment and net exports. Let us try to annualize the quarterly growth rates in both real and nominal terms which yields a quite different picture. In real terms Japan thus grew 1.8% through 2007 and in nominal terms we are down to a rather un-impressing 0.6%. Particularly for Q4 the figures are 0.9% and 0.3% for real and nominal growth rates respectively. In this way, the GDP deflator is a welcome alternative to the CPI index in that it accounts for the change in prices relative to what consumers actually buy in the measurement period. Thus note in passing the following from Bloomberg …
Rising oil prices may have boosted growth in real terms. The GDP deflator, a broad measure of prices used to calculate real growth from nominal, fell 1.3 percent from a year earlier, the biggest drop since the first quarter of 2006. The deflator is adjusted downwards when oil prices rise. In nominal terms the economy grew an annual 1.2 percent in the fourth quarter.
I am going to discuss this more below since inflation measurement is clearly one of those areas where data mining and basket building can be used to construct just about any kind of number you would like. In this way, all these kinds of inflation adjusted growth rates etc need to be taken with a pinch of salt. In conclusion on the GDP figures I think the following is important to note. First of all, this is good news since it indicates, all things equal, that Japan has defied at least some of the claims that a recession/slowdown is imminent. However, I am not sure how much valuable information we can reasonably derive from the figures at this point. First of all, I think these figures are in for a haircut once they are subject to revision. Yet, even if we rely on them such as they are I think that it is reasonably clear how for example the value component of energy prices might have pushed up the real GDP to unrealistically high levels if we consider the underlying trend.
2) Inflation: see my previous posts: CPI is just a price index. Its definition is very different from a country to another. Change in the index & methodology would give a total different picture. Most important is the trend of the index itself, and the recent trend is up. It’s ridiculous to speak of deflation any more. CPI has been ranging from -2% to 1% in last years. It doesnt make a lot of difference. Asset-prices are more important: Nikkei is still nearly more than 80% off 2003 lows, real estate in central Tokyo as well despite being off its highs probably by 20% at least.
Unfortunately I am not sure about which posts Francois is referring to here but nevertheless he fires off a lot of reasonable questions here. First of all, I completely agree with the point on methodology. Since the CPI is based on a basket which can be changed and re-weighted and since the CPI may or may not include headline inflation what we end up with is a veritable mind field of potential ‘best practices’. This also means that whatever the picture you want you can rig the data so that your specific view of the world emerges. I don’t think however that this is what I have falling victim to in my analysis of Japan. In my opinion, the price movements in Japan both in the most recent period as well as in 4-5 year perspective show two things. First of all, there is the overall level of inflation which has been very low and essentially negative. This, coupled with the very aggressive monetary regime put in place to normalize conditions indicates I think that there is indeed ‘something funny’ about consumption and domestic demand in Japan and this is what has led me to conclude that the whole price edifice in Japan has something to do with the population structure of the country. Secondly and in the more immediate context the recent divergence between input and output prices further support my claim that price dynamics in Japan do not follow the theories we can discern from macroeconomic textbooks and traditional empirical studies. Moreover, it obviously suggests that the equality often exclaimed in the financial press between the return to inflation and economic recovery is wrong.
Now, all this leads me to disagree with Francoise when he says It doesnt make a lot of difference. I think it does although I do agree that asset price deflation/inflation is extremely important too. In this respect the Nikkei is mentioned being considerably off its low levels of 2003 as well as those much debated Tokyo real estates have seen hefty increases in price. Both these points are very important to take aboard I think and merits, at least a bit, that Japan has moved on. Of course, the most recent developments in the Japanese housing sector suggest that the construction/residential sector in Japan might very well be in for a more difficult future but let us leave this point here. However, what about another kind of asset in the form of human capital? How does the value of this asset stand? Well, as we have observed one of the recent trademarks of the Japanese labour market has been a consistent decline in wages and the transition from a labour market of full time workers to part time workers (on the margin of course). Since aggregate national wages essentially can be seen as a measure/reflection of the national labour productivity (either absolute or per/hour) what does this imply for the general price level in Japan? As can be seen, this readily becomes rather complicated. Another reason as to why inflation matters has to do with the workings of a modern economy is monetary policy. Quite simply, deflation or next to no inflation has implications for the workings of monetary policy as well as it has implications for the consumption dynamics of the society. More importantly, we have seen the condition of deflation in Japan and subsequent low interest rates have had notable externalities on the global economy. So, I would say that it does indeed matter.
3) Monetary policy. I fear there could be a big misconception on USD buying interventions. MOF as you know is running a hell lot of debt. It is short Yen cash. But it seems to me that most of the FX intervention is executed by BOJ, but on behalf of the MOF. So when MOF buys USD, it needs to borrow even more JPY. But with the end of Quantitative easing, they can’t issue as many Financial Bills to back them as they would like to (because BOJ would basically buy an unlimited amount of them @ 0% before.) I think that with the deterioration of public finances, it becomes harder to do such intervention. So I see just a lot of talk, not much more.
Here I stand corrected. Consequently, I had not, in my analysis of the JPY and subsequent potential for intervention, thought about the perspective Francois presents here. It is very interesting I think. Now, for some of our readers this may seem a bit complicated but what Francois is saying is simply that absent quantitative easing/ZIRP it becomes more ‘expensive’ for the MOF to intervene since they cannot be sure that they are able to offload the subsequent debt. Of course, this also paves the way for a rather perverse scenario. Consider thus that the JPY is driven largely by risk sentiment at the moment. If the BOJ sees it fit to lower rates during the course of 2008 and perhaps even returns to ZIRP we could expect the JPY to shoot up given we accept the current market dynamics. Note in passing here that this morning’s GDP release has been followed by a depreciation of the JPY which shows the disconnect between the fundamentals and the JPY. In this way, a return to ZIRP or just a drop to 0.25% could in this context be followed by an increased in the pressure to intervene. Of course, this is not a plausible scenario at this point but still goes to show the potential dynamics as we move forward.
4) JPY everybody I talk to is bullish on the JPY... maybe that is why it is so sticky now...but its way off its lows against many crosses.
As I have said above and also in my recent review and preview the JPY remains wholly disconnected from the fundamentals of the Japanese economy. I concur with Francois that the sentiment on the JPY at the moment seems to be bullish given the general risk sentiment in the markets. At time of writing it is sniffing at 108+ which is outside the recent weeks’ range of 106-107. It is difficult to see where it goes from here. I am expecting this ‘stickiness’ theme to dominate since it is unclear I think whether market conditions would favor a move below 105 or upwards to 110.
5) Long term interest rates... the credit markets have imploded in less than a year. My prediction is for a failure of a big govt bond market in 5 to 10y time. Japan would be an obvious candidate. In that scenario, long term interest rates are heading HIGHER. Just people will be tired to be stuck with low interest rates when there is inflation everywhere. But in the short term, as the asset-bubble is deflating, and this process is not over, global govt bonds will remain for some time the asset of choice, by default.
Now, this is very interesting in my opinion. Whether or not we will see a failure in a government bond market is an open question subject to one of those rather long term falsification clauses. I have argued before that in the context of countries such as Japan and Italy it will, at some point, cease to make sense in ‘rating’ the sovereign debt market based on the same criteria as you treat e.g. India, the US etc. Quite simply, this will become unfeasible as we move forward since this would push these countries into a technical default. As Francois alludes this may of course come to pass some way or another not because of the rating agencies themselves but rather because with inflation the nominal yield may become too unattractive. Note also that once we enter this discussion we also enter a whole gamut of issues in the context of Japan in the sense that the BOJ and the MOF is in a double bind. On the one hand the BOJ faces external pressure (those externalities again) to raise or more aptly to normalize rates but it finds this difficult because deflation still dominates the general price level. Moreover, a transition towards whatever the interest level we assert to be normal would most likely drive up the value of the JPY and thus further lead to deflationary pressures. We should also consider the simple points that as interest go up the debt becomes more expensive to service and in this way the MOF has a distinct interest in keeping interest rates down. Within this framework headline inflation pressures are of course simply a further pinch it seems not least because of the reasons mentioned by Francois.
This topic on sovereign debt and long term interest rates is very important I think but for now I think that I will lower my guns. Thanks for Francois for the comments. Needless to say, here at JEW (and at Alpha.Sources) we always appreciate to be challenged on our views and opinions.