Follow the Money

Maybe it is just me, but Thankgiving 2006 is starting to feel a bit like the period after Christmas in 2004.   Both periods saw sharp falls in the dollar in thin markets.  1.31 isn’t 1.36.  But, in a (almost) no volatility world, a sudden move to 1.31 certainly generated headlines. I certainly don’t know if the dollar will rebound when more normal market conditions return next week.    Some certainly think so.  But there are no shortage of reasons why the dollar could stabilize below 1.30 – a slowing US economy, smaller growth (and interest rate) differentials between the US and Europe and that large (and still rising) US current account deficit.  BNP Paribas seems to be among the structural dollar bears:“To dismiss this as a technical correction is to overlook the structural reasons why the U.S. dollar is having a very hard time these days,” said Hans Redeker, global head of currency strategy at BNP Paribas. Some carry traders somewhere must be nervous.   At the Euromoney conference in early November, pretty much everyone on the program seemed to expect that the low volatility environment that makes carry trades attractive would continue.High carry strategies (at least strategies that involved borrowing low-yielding G-10 currencies to invest in high yielding G-10 currencies) have made money nine of the last ten years.  1998 is the only exception.   That was one thing I learned at the Euromoney conference.Most expected “high carry” strategies to continue to do very well – not just in 2006, but also in 2007.    I don’t think borrowing euros to buy dollars has been a popular carry trade.  There isn’t much carry relative to the risks – particularly as the euro has rallied against the dollar this year despite somewhat lower eurozone rates.  But borrowing yen to buy dollars certainly has been a reasonably popular high-carry strategy.    And the yen joined the euro in this week's move, even if it doesn't seem to have moved as much on Friday.  For all the parallels with 2004, though, there are no shortages of differences between the fall of 2004 and the fall of 2006.Here are the ones that stand out to me. China is far more exposed to a big fall in the dollar now.    China has twice as many reserves today as it had two years ago.   It may not quite have twice as many dollar reserves as it had two years ago (I suspect it diversified a bit during the dollar’s rally v. the euro in 2005), it still has a lot more dollars than it had a few years ago.   China’s dollar reserves are almost north of 25% of China’s GDP ($700b v. $2.6 trillion or so).   There was an active debate inside China about China’s potential (over) exposure to the dollar even before the dollar’s recent slide.   But it isn’t clear to me if that has generated a consensus on what to do.  China has already tried most of the easy options.    Allow a bit more flexibility.   Done that. Keep Chinese interest rates below US rates.  Done that. Make sure the RMB’s appreciation is less than predicted in the forward market.   Done that. Loosen controls on capital outflows.  Done that.  Increasingly allow firms to keep their dollar export proceeds on deposit in the banking system in dollars rather than convert them into RMB.  Done that. Convince the banks to hold the funds raised in their IPOs offshore.  Done that .... despite all these efforts, China’s reserves are still rising by $20b or so a month, and no doubt it dollar holdings are rising fast.   The RMB’s slide v. the euro won’t help China’s efforts to rebalance its economy away from exports either. The oil exporters are much more exposed to falls in the dollar than in 2004.    They have stuffed a tremendous number of dollars away over the past few years.   Even those that have shifted their portfolios toward euros (Russia) have more dollars than they did a few years ago – simply because they have a lot more money to invest in international markets.   The Saudis presumably have even more dollar exposure … GCC has even less need for a weaker currency right now than in 2004.   Back in 2004, I think most of the oil exporters were expecting oil to fall back and budgeting very, very conservatively.   The GCC still has a huge cushion between budgeted spending (inlcuding spending on investment projects) and export revenues.  But there also is a fair amount of additional domestic spending (including spending on ‘investment projects”) in the pipeline.  Indeed rising spending -- including a surge in construction -- led GCC inflation to pick up even before the GCC currencies joined the dollar's most recent slide.  But if China and the Gulf are more exposed to a fall in the dollar than they used to be, the same cannot really be said of the rest of Asia.  In late 2004, most emerging Asian economies had been intervening almost non-stop for several years.     The dollar’s 2005 rally let them get out of the market (and no doubt helped many diversify).  They were active in the market sporadically in 2006 (including, I would bet, on Thursday and Friday), but they have been intervening on a sustained basis.     As Stephen Jen notes in revising his euro/ dollar forecast for the end of 2006, Europe’s economy looks healthier now than it did in 2004.  The euro’s rally v. the dollar in late 2004 came in the face of rather sluggish European growth.   Somewhat ironically, the euro area economy picked up in 2005 amid a big bout of euro-pessimism in the currency markets (and all sorts of political angst).    And no doubt some European finance ministers are worried that the euro’s current strength will trigger a renewed bit of sluggishness in the eurozone.    Think the RMB is weak v the dollar?  Look at the RMB/ Euro ...The US needs rather more financing than it did in 2004.   The US current account deficit in q3 2006 is likely to be around $900b annualized … and unless the US starts cutting rates, I suspect the current account deficit will continue to rise in 2007 even if the trade deficit stabilizes or falls, thanks to a rapidly rising interest bill. One thing though probably hasn’t changed much.  The US net international investment position -- the gap between the dollar value of US external liabilities (FDI in the US as well as US borrowing from the world) and US investment abroad (inlcuding US lending to the world).   Stock markets outside the US generally have done well again this year.   And the euro’s slide increase the dollar value of US investment in Europe.   The rising value of US external assets should help to offset all the debt the US is taking on.  Nothing beats borrowing against the rising value of your external assets.One of the features of the international financial system over the past few years is that any fall in the dollar has been met by both an increase in the overall pace of reserve growth (countries that peg to the dollar have followed the dollar down, and many countries have intervened rather than allow their currencies to move up) and an increase in the share of that reserve growth that is held in dollars.   So a weaker dollar generally has meant more rapid growth in central bank dollar reserves: central banks have financed the US when the markets don’t want to.At the end of 2004, I suspect some central banks had rather more dollars than they wanted.    Many were able to shift in euros over the course of 2005 – that was one side effect of the Homeland investment act and the (failed) referendum on the new European constitution. Other central banks though continued to pile up the dollars.  China and the oil exporters most notably.If the dollar’s current slide continues, those countries face a set of increasingly difficult choices. China’s current (not-a-real basket) peg implies that it would need to pick up its dollar reserve accumulation to keep the RMB from rising against the dollar, even as its central bank talks of diversification.The GCC countries would be faced with a similar set of choices – their currencies are depreciating in real terms as well.     That will only add to (strong) inflationary pressures in the most rapidly growing GCC countries – and force the GCC countries to choose between more sterilization (which likely means scaling back spending plans), higher inflation (and real appreciation from faster price rises than in the US) or a revaluation (and a real appreciation from a nominal appreciation). And emerging Asian economies (from India to Korea) could have to choose between allowing their currencies to appreciate (or, for some, appreciate more) against the dollar and renewed large-scale intervention.No doubt, there are lots of folks hoping the dollar rebounds on Monday.  So long as the dollar stays in its recent ranges, many key actors can continue to postpone some increasingly difficult choices.  But if the dollar slides, some have to choose whether or not to join the dollar on its way down .. and others have to choose whether or not to join the euro and the pound on the way up.

I have outsourced Thanksgiving blogging to Charles Gottlieb of the Center for European policy studies in Brussels.   (Charles.gottlieb at ceps.be)His topic: The Spanish and Irish housing booms (or bubbles).   The Spanish and Irish economies are even more housing-centric than the US economy ... and potentially are even more exposed to a housing slump.  Enjoy!  Red alert in the Euro zone-periphery – some are still riding the housing bubble As argued all along the housing euphoric “literature”, global factors have greatly fuelled housing prices in Europe.  The historical lowness of interest rates has played a great role, but also the development of financial systems that allow people to borrow against their future income and the home’s value.Yet even amid a global housing boom, Spain and Ireland stand out. France –which has experienced very strong home price appreciation recently seems bound to cool (see my previous contribution).  But Ireland and Spain are still rocketing, with home price growth of around 10% yoy (INE and CSO) over 2006 … (UK too, though after a pause).  These two EU periphery countries exhibit the biggest housing price hikes, highest residential investment (relative to GDP) and most jobs in construction sector.  They are consequently among the most housing-centric economies in the entire world …. Spain and Ireland have benefited from strong convergence-related growth and positive population dynamics.  But it now seems clear that those fundamentals don’t suffice to justify their housing prices dynamics.   Economic agents are increasingly exposed to interest rate hikes, and the economy as a whole is increasingly tied to their construction sector. Hence Spain and Ireland have ridden the housing boom more than most, with better than average economic performance… but now a red alert looms.What explains the biggest housing booms in the OECD? Ireland and Spain’s over-muscled fundamental Spain and Ireland both greatly benefited from low interest rates, from favourable migration dynamics, and their housing sector has both benefited from (and contributed to) strong local economic growth. By importing the credibility of the European Central Bank, they benefit from low nominal interest rates, in spite of their vivid growth and consequent higher than average inflation. Thus in addition to the global savings glut which exercised downward pressure on nominal interest rates, both countries exhibit very low real interest rates. Up until European monetary integration, both countries used to be European outliers. Both countries had not benefited from the European post-war wealth surge, and had living standard well below the European average. However, membership in the European Union and EMU triggered a strong “catch-up process”.  On its current path, Spain will have fully converged with the Euro zone’s big three (France, Germany and Italy) in only 7 more years. Such development have unleashed “animal spirits,” made both countries attractive destinations for immigrants (a big historic change), and laid down a strong basis for their real disposable income growth. Simultaneously, the perception of increased economic certainty, changes in financial markets and financial innovation in Spain and Ireland have made mortgage credit more available. The deepening of mortgage markets has helped to sustain the unprecedented persistence of demand for housing amid surging home prices. In fact, the total value of mortgage debt in Ireland tripled over the 2000-2005 horizon, and mortgage lending is still growing at a 20% pace.  Spanish lending is still growing at a 23.6% pace. Given that in Ireland 83% of total outstanding mortgage debt in 2005 and in Spain, 97% of the debt is at variable rate interest, households are considerably exposed to interest rate hikes. The sensitivity of both countries’ market to interest rate change is thus considerable, and could rapidly trickle down to households’ balance sheets as monthly repayments are bound to rise (considering the current lowness of interest rates in the Euro area). Also a sharp increase in interest rates may trigger reduction in housing demand, raise the repayment service of indebted households and could backfire onto the financial system. Already in Spain, the average person with a mortgage allocates fifty five percent of their wage on their principal or secondary home repayments a share which increased by 4.1% since 2003 (La Caixa, 2005).Both Spain and Ireland are more American than the US: they have comparable population dynamics (thanks to migration) and faster growth in both real disposable income and real housing prices.   Chart 1: Growth over the 1999-2005 period    Source: OECD database, Eurostat.However, Spain looks a bit frothier than Ireland.   Spain has lower real income growth than Ireland but faster home price appreciation.  Spanish nominal wage increases have remained above the euro area average despite low productivity gains.  Amid strong demand pressures and weak competition, companies have been able to pass on relatively rapid labour cost increases into prices.   And strong wage growth, in turn, has helped to support the surge of housing prices (this works both ways: given the labour intensive character of the housing industry, higher wages have pushed up new home prices). Home construction costs have risen by 25% for Ireland and 33% for Spain between 2000 and 2005, contributing to the 64% and 84% respective increases in home prices over the 2000-2005 period. The surge in prices in both countries does not reflect limits on supply:  unlike in the UK and France, investment in the residential sector in Spain and Ireland has boomed.   In Ireland investment in housing represents 14% of GDP, and Spain 9%, while most other Eurozone countries exhibit a share of 5% of GDP (Eurostat 2005 data).  (Residential investment in the US peaked at 6.2%).Consequently, both countries are increasingly exposed to a slump in investment even if it doesn’t spill over to slump in consumption.  Spain and Ireland also are at the top of the European scale when it comes to employment in construction, with home construction accounting for approx. 12% of employment in 2005. Construction output per capita in Ireland is highest relative to Euro zone countries at approximately €7,600 and house completion was four times the average of other European countries in 2005 (CSO report), twice  the corresponding figure for the UK.   The National Bank of Ireland recognizes that a sharp reduction in housing output would lower employment, investment and growth.   But the exposure of the Irish economy to a slowdown of the housing sector seems even large than the monetary authorities admits. Too lenient tax incentives have caused housing price overvaluation and lured too many resources towards residential investment. This has lead to an inefficient allocation of resources, which could particularly hamper Irish and Spanish growth going forward.A reversion to mean in the housing sector would imply a big fall in all economic measures, and might introduce a temporary recession in Europe’s periphery. As emphasized in my first post, the wealth effect from a slump in housing prices is considerable in Spain and Ireland. However even if these indirect effects reveal to be more contained, direct effects through investment and employment in the housing sector will curb growth and necessitate considerable reallocation of resources away from housing.  Sociology  intertwined with EconomicsThe housing markets in both Ireland and Spain have been supported by a recent influx of migrants’. Approximately 54 per cent of new immigrants belong to the 25-44 age categories, the age category most likely to invest in housing. But migration matters less that domestic social factors – and national policy choices.  Spain and Ireland have -- with Italy -- the highest home ownership rates in Europe (Spain 85%, Ireland 77% and Italy 80% as of 2002) as a result of long term national policies favouring of house owners.   In Ireland, tax treatment of housing is very favourable for home ownership compared to other EU countries (van den Noord, 2005). In Spain, any housing property is tax-deductible and for every sort of income source, whereas ordinary rental properties are not.  Such asymmetrical treatment of rental compared to property has lead to high ownership rates (see BIS paper). Such regulatory measures impeded the emergence of a rental market. Indeed, Spain (and Italy)’s population relies on a traditional family structure to compensate for lack of rental opportunities, infrastructure, and low public transfers for the young. Young Spaniards and Italians face the so-called Delay Syndrome; 63% of 20-34 year old unmarried co-habited with their parents in Italy. In Spain, the equivalent figure is 40% as of 2004 (compared with 20% in Germany).  The average size of Spanish households, while falling, is the highest in Europe (INE stats).  A more differentiated housing market which allows for all age categories to choose between living at home, renting and owning would be better than the status quo.On the one hand, the lack of rental market in those countries inhibits the Spanish youth to gain independency from their parents. On the other, the house owners refraining from renting their housing property are this generation’s parents… To develop Spain and Ireland’s rental markets fiscal incentives favoring owners have to be abolished or at least rebalanced to the benefit of tenants. In Spain an institutional setup has to be implemented that secures alternatives for renting out owned housing, so that owner’s don’t need to occupy it. Such measures would reduce the amount of vacant properties which in Spain amounts to three million properties, and ease housing prices.Red alert in the periphery Central banks are concerned that recent pursuance of housing price growth in both countries wasn’t supported by fundamentals. The Irish national Bank stated in its latest financial stability report that the 2006 price surge wasn’t expected. In Spain, the Central bank has already issued some warnings regarding credit risk monitoring. The IMF Directors noted “that an abrupt correction cannot be ruled out” in Ireland.  Cotis from the OECD has acknowledged that several big countries are at risk of a housing downturn: with the USA, France and the UK topping the list.  But, given the extreme dependence of both Spain and Ireland on housing, both countries are even more exposed to a sharp correction. And there is one key difference between the small – and no longer so small countries like Spain -- countries on the periphery and the bigger economies: just as the ECB hasn’t raised rates to offset the boom in home prices in Europe’s periphery, it is unlikely to lower rates just to help smaller economies.  Wine and roses don’t last forever.  Monetary union has offered huge advantages to both Ireland and Spain, but it won’t always generate the favourable dynamics both countries have enjoyed recently.

One thing consistently surprises me: how little most folks in the US markets – including folks active in international markets -- know about the growth in reserves of the oil exporters.    China, everyone gets.     But not Russia and Saudi Arabia.   Estimates of oil reserve growth at the Euromoney fx conference in New York were stunningly off (and way on the low side).   And there is a lot less talk of petrodollars than Chinese dollars these days.  There are reasons for this.  China buys directly from the US broker-dealers.  It is on track to buy over $100b or so directly from the broker-dealers this year (as well as building up its short-term claims).   No doubt its total purchases are higher. The oil exporters tend not to buy directly.   They buy through intermediaries, build-up bank accounts in London (which help finance London hedge funds and others buying riskier US assets) and invest directly in hedge funds and other money managers.  They aren’t as visible a presence, at least in the US.    Russian reserves are up $94.8b through mid-November.  Setting the second quarter aside (for seasonal reasons), Saudi Monetary Agency foreign assets are growing by a bit less than $25b a quarter, so its total reserve growth for the year should top $75b.  That's real money.   The Emirates doesn't report how much it is putting into its various oil investment funds, but $40-50b seems reasonable.   And so on. Some oil exporters are so conservative that they haven’t been a big player in the debt market -- Russia, for example, has a very conservative portfolio of portfolio of short-term agencies and bank deposits.   Other oil exporters hold a very diverse portfolio – one that includes equities and emerging market debt, reducing their impact on the US fixed-income market.  ADIA (Abu Dhabi’s Investment Authority) is a good example.   No doubt some oil exporters also farm out management of some of their portfolio to US and London and other fund managers, effectively financing a lot of “private” market activity. As a result, it is hard to find oil related flows in the US data (this RGE proprietary paper has the details).   It is also hard to find all Chinese flows – but it is easy to find about ½ of them. Still, it is worth remembering that the Gulf (the GCC countries) will have a current account surplus that is only slightly smaller than China’s surplus this year.     Only about ½ of that surplus goes into formal reserves (counting all SAMA foreign assets) – the rest goes into oil investment funds.    So GCC reserve growth lags Chinese reserve growth.   But if you combine Saudi and Russian reserve growth, it is comparable to Chinese reserve growth …  but it also attracts a lot less attention. Two graphs illustrate the impact of oil exporters on the global balance. The first comes straight from the IMF.  It compares the expected surpluses of the major oil exporting regions (The Middle East, Russia and Africa – think Algeria and Libya as well as Nigeria and Angola) with the expected surpluses of emerging Asia (including the NIEs) and other emerging economies (Eastern Europe/ Latin America).    The measure of oil exporters misses Venezuela, so it isn’t perfect … and the IMF’s calculations use an average oil price that looks to be about $5 too high for 2006, so its estimates are likely on the high side for the oil exporters and the low side for Asia.  Enough throat clearing.  The picture speaks for itself: oil exporters are currently the dominant source of the emerging world’s current account surplus.   The second graph compares the gulf’s (GCC countries – Saudi Arabia, the Emirates, Kuwait, Qatar, Bahrain and Oman) cumulative current account surplus since 2000 with China’s cumulative surplus.    China’s cumulative surplus lags the Gulf’s. China’s cumulative reserve growth leads the combined growth of GCC reserves and the increase in the GCC’s oil investment funds (at least that portion of the increase that comes from new funds, not capital gains), as China has attracted large (net) capital inflows.    And now that China has a $230-240b current account surplus, it is on track to build up its current account surplus more rapidly than the Gulf. But it is still impossible to make the global current account balance – or to find out how the US has financed its “almost inconceivably large” current account surplus -- without taking into account the oil windfall.    But absent more visible flows, lots of folks don’t recognize this.