Singapore
Upgrading 2007 Forecast
April 11, 2007

By Deyi Tan and Chetan Ahya and Tanvee Gupta | Singapore, Mumbai

1Q07 advance estimate at 6.0% YoY

The Ministry of Trade and Industry estimates Singapore’s 1Q07 GDP at 6.0% YoY, compared with 6.6% in 4Q06.  This is higher than our and consensus forecasts of 5.0% and 5.5% YoY, respectively.  While the overall growth trend continues to be one of deceleration, we believe that the government’s policies to diversify risks and reduce dependence on a particular sector appear to be paying off.  For instance, the continued strength in transport engineering has helped partially offset the adverse impact of a sharp slowdown in electronics and the chemicals sector.  As a result, deceleration in manufacturing has been relatively moderate.  It is estimated to grow 6.1% YoY in 1Q07 compared with 7.7% YoY in 4Q06.  Similarly, the services sector is somewhat supported by the government’s initiatives in nurturing other segments such as wealth management and is expected to grow at 6.1% YoY in 1Q07 (versus 6.6% YoY in 4Q06).  The overall growth trend has also been supported by continued acceleration in the construction sector.  In 1Q07, growth in construction is estimated to have improved to 7.0% YoY from 4.7% YoY in 4Q06.

Upgrading 2007E growth from 5.0% to 5.5%

We are raising our 2007 GDP growth forecast from 5.0% to 5.5% and our 2008 GDP forecast from 5.5% to 5.7% to reflect the relatively stronger-than-expected underlying domestic demand growth.  First, as we noted in First Signs of Recovery for Low-Income Groups, March 2, 2007, the lower-income deciles are finally beginning to see an improvement in income growth, albeit at a slower pace.  We believe that this should help support private consumption.  However, note that we expect this improvement to be relatively small.  A large part of the income growth in the low-income deciles has been due to job creation rather than wage growth.  Second, the recovery in the construction sector appears to be firming up.  Construction work for the integrated resorts commencing in 2007, as well as for office space and residences from the property upswing now under way, should continue to support overall growth going forward.

Positive monetary policy support

Despite healthy growth and a tight labour market, the inflation front stays benign at below 1%.  This is on the back of the appreciation bias in the currency, which helps to keep import prices down, and falling utility and Certificate-of-Entitlement (COE) prices.  Inflation ex-private transport, which strips out movements in COE prices and oil prices, has also been decelerating, staying at slightly in excess of 1%.  The 2% Goods and Services Tax (GST) hike, which will kick in in July, will pose some near-term upside risks, in our view.  About 75% of the basket will be affected by GST changes, according to a report by the Monetary Authority of Singapore put out in 2003 when GST was to be raised by 1%.   However, there is unlikely to be full pass-through, judging from past instances of GST rises, and since retailers might absorb some of the costs.  Taking into account the recent movements in inflation and the 2% GST hike, we are raising our 2007 inflation forecast from 1.0% to 1.1% and our 2008 inflation forecast from 1.2% to 1.4%.

External risks persist

Despite increasing diversification of business risk, the influence of the external sector remains strong.  Apart from the direct impact of export growth, the second-round impact on export-oriented industry’s capital spending tends to influence the overall growth trend.  External demand from major markets such as the US, the EU, Japan and China is slowing. Machinery and equipment linked to the export cycle will likely follow suit.  The US ISM New Orders Index, which leads NODX by about 1-2 quarters, is signalling a slowdown that has yet to bottom out.  Currently, our US economics team expects GDP growth in the US to start recovering from 4Q07.  The possibility of an extended period of slowdown in the US remains a risk factor to our growth estimates for Singapore.

Japan
Falls in Leading and Coincident DIs a Portent of Recession?
April 11, 2007

By Takehiro Sato and Takeshi Yamaguchi | Tokyo

Implications of the recent decline in leading and coincident indexes

The coincident index issued by the Cabinet Office came in below 50% for the second straight month in February, and all three indexes (leading, coincident and lagging) had sub-50% readings for the first time since December 2001 at the final stage of the economic recession following the IT bubble. Leading and coincident indexes were both below the 50% mark for a second straight month in February for the first time since November-December 2004, when the economic activity flattened from IT adjustments after the Athens Olympic Games, and pessimism weighed on sentiment. Some investors view the downturn in business conditions indexes as a sign of an economic pullback. This stance may appear to have some merit, simply judging from the past pattern.

Yet we disagree with this conclusion and attribute recent weakness to a combination of special factors. There are past examples of a similar pattern and the economy did not always proceed to a recession. Below we review special factors affecting index components and economic performance in previous cases of leading and coincident indexes posting two straight months of sub-50% readings.

Special factors in the leading index

The leading index was 30% in February (preliminary data). The index of producers’ inventory ratio of finished goods, TOPIX stock prices and sales forecast DI for small businesses improved from November. The other seven components (index of producers’ inventory ratio of finished goods, new job offers, total floor area of new housing starts, index of producers’ shipments of durable consumer goods, consumer confidence index, Nikkei Commodity Price Index (42 items) and the interest rate spread) weakened from November. We think that new job offers, the interest rate spread and the Nikkei Commodity Price Index probably declined for reasons other than the economic cycle.

(1) New job offers: Companies restricted new offers in response to tougher scrutiny from authorities toward fake subcontracting practices and falsified job openings. While new job offers are a leading indicator for effective job openings and typically move ahead of economic activity, the latest decline is unrelated to the economic cycle, in our view. In fact, the BoJ’s March Tankan survey reports growing labor shortages at companies of all sizes.

(2) Interest rate spread: The interest rate spread narrowed in February from the lack of a reaction by long-term bond investors to the BoJ rate hike. However, recent stability in the bond market reflects non-fundamentals factors such as a structural decline in demand for funds amid rich corporate cash flow, and real-money investors including life insurers/investment trusts increasingly selecting long-term bonds due to excess demand for duration. We refer to this trend as a revival of the cash flow dynamic. The yield curve has also flattened globally unrelated to the economic cycle. We conclude that the yield curve shape does not necessarily provide an accurate indication of future economic activity.

(3) Nikkei Commodity Price Index (YoY change): Momentum eased on a YoY basis even though crude oil and other primary product prices slowly increased. Yet it is natural for the YoY trend to weaken with primary product prices trading at historically high levels. We also think that monthly movements of primary product prices do not necessarily correlate to the economic cycle, given increased volatility amid global excess liquidity.

Special factors in the coincident index

The coincident index was 16.7% in February (preliminary data). The index of non-scheduled worked hours was the only component posting an improvement from November. Seven components weakened (index of industrial production, index of producers’ shipments, large industrial power consumption, index of sales for SMEs, index of producers’ shipments of investment goods, wholesale sales value and the job offers to applicant rate) and retail sales value was flat. We think that the job offers to applicant rate and production-related indexes probably declined for reasons other than the economic cycle.

(1) Job offers to applicant rate: The job offers to applicant rate peaked at 1.09x in July 2006. Yet the downward trend can be attributed to the same factors as the new job offers rate and is somewhat misleading. We place more emphasis on growing labor shortages indicated by sentiment data (BoJ Tankan and Economy Watchers surveys) as mentioned above.

(2) Production-related indexes: Industrial production strengthened in October-December on robust momentum from the automotive industry with a healthy 2.6% QoQ gain. We expect a reversal decline in January-March for the first setback in six quarters due to modest inventory adjustments mainly for automobiles. Yet the modest inventory adjustments should not last very long. The manufacturing output forecast survey for March and April is fairly upbeat at +1.5% and +1.3% MoM, respectively, even though IT-related inventories have been rising to unprecedentedly high levels. We are obviously not counting on a full achievement of METI’s production forecast, but still project a moderate overall recovery by industrial production in April-June. This view applies to the other production-related indicators such as the index of producers’ shipments, large industrial power consumption and the index of producers’ shipments of investment goods as well. Investment goods shipments, a coincident indicator of capex, should be relatively healthy, considering upbeat capex plans mainly at non-manufacturers reported in the BoJ Tankan survey.

Review of past patterns

Next, we review the relationship with the actual economic cycle when leading and coincident indexes were below 50% from past data.

(1) Two straight months of sub-50 readings for the coincident index: The coincident index has posted sub-50 readings for two straight months on two previous occasions in the current economic expansion phase since February 2002 (October-December 2004 and March 2006). We find 13 cases of sub-50 readings for two straight months over a longer period from April 1980 to December 2006, with four during economic recoveries. The coincident index hence does not always coincide with the economic cycle and has a miss rate of 31% (4/13).

(2) Four straight months of sub-50 readings for the leading index: The leading index has posted sub-50 readings for four straight months only once before in the current economic expansion phase since February 2002 (December 2004). We find nine cases of sub-50 readings for four straight months during April 1980 to December 2006, with economic recessions in the next six months in six cases. Economic recovery continued in the other three cases. This puts the miss rate at 33% (3/9).

(3) Two straight months of sub-50 readings for the leading and coincident indexes: The leading and coincident indexes exhibit close correlation and we find 13 cases of sub-50 readings for both indicators for two straight months during from April 1980 to December 2006. Economic recessions occurred in the next six months in nine cases, while economic recovery continued in four cases, putting the miss rate at 31% (4/13).

The miss rates should be interpreted broadly since results vary considerably depending on the period and sample count. Leading and coincident indexes also pick up smaller economic waves that are not covered by the official economic cycle. Some ‘miss’ cases stem from unofficial modest recessions or plateaus (such as the first half of 1995 and second half of 2004). Yet we advise against mechanical conclusions about economic recessions based on index movement, since these indexes do not have extremely high accuracy in predicting major economic cycles.

Our outlook

We think that investors will have difficulty shedding worries about the US economy. In fact, our US economics team has lowered the forecast for US economic growth in 2007 from 2.4% to 2.0%, despite healthy jobs data in March, reflecting further adjustments in the housing market and recent sluggish capital investment. The coincident index must consistently stay above 50% to remove investor worries under these conditions. This will take a bottoming out of the job offers to applicant rate and other employment-related indicators and a recovery in production activity and capex-related indicators from April-June. We basically maintain our optimistic stance, as explained above.

The late 2004 case, meanwhile, was linked to flatter economic activity that did not reach the level of an official recession. With regard to this, we anticipate nearly 3% annualized growth by the Japanese economy in January-March 2007, aided by a further recovery of personal consumption in contrast to the downturn in recent business conditions index data; however, the April-June growth rate might ease to the lower 2% range, despite a rebound in manufacturing output, in a reaction to strong October-December and January-March trends and reflecting the slower US economy. That said, our economic forecast already discounts to some extent for slower momentum in April-June and July-September.

Israel
The Poverty of the Welfare State
April 11, 2007

By Serhan Cevik | Istanbul

Breaking away from the legacy of welfare state, Israel has achieved faster income growth. Two factors — the lack of political direction to resolve territorial conflicts and the burden of an ever-growing welfare state — have kept the Israeli economy below its true potential. Even though a comprehensive peace deal with the Palestinians still remains in a distant future, structural reforms and prudent fiscal management in recent years have helped to rationalize Israel’s welfare system and thus reduce its burden on the economy. With intensifying political pressures over the decades, the cost of welfare expenditure and transfer payments to households increased from 31.5% of GDP in 1980 to 42.2% in 2002. As you would have thought, such an extravagant level of social spending not only worsened public finances, but also distorted economic incentives and lowered the economy’s growth potential. Realizing the extent of structural strains, especially on income growth in the private sector, the authorities have finally started moving away from the legacy of welfare state and dealing with institutional constraints. Accordingly, the burden of welfare spending and transfer payments declined rapidly to 36.5% of GDP, helping to lower the overall budget deficit to 0.9% of GDP at the end of last year and giving a significant boost to the economy.

Rationalizing welfare expenditures has merely unveiled the ‘hidden’ poverty. Israel has enjoyed robust output growth and the lowest unemployment rate in a decade, but the benefits are not distributed equally, at least in the early stages of normalization. Consequently, the number of people living below the relative poverty line increased from 20.6% in 2002 to 24.7% in 2005. Although the rationalization of welfare spending has contributed to diverging growth rates across the income spectrum, it merely unveiled the ‘hidden’ poverty. Indeed, excluding transfer payments and taxes, the poverty rate was already much higher throughout the 1990s and even showed some improvement in recent years. Moreover, defining the poverty line as half of the median income after transfer payments and taxes distorts poverty statistics. Since the upper quintiles of the income spectrum have experienced faster growth in the initial phase of economic expansion, it is normal to see an increase in the relative incidence of poverty. Nevertheless, the rationalization of social expenditures has already led to a small but encouraging improvement in the relative poverty rate to 24.4% by mid-2006.

Institutional factors and structural changes are behind Israel‘s poverty problem. Fiscal stabilization measures and reforms lowered the number of people receiving income support and unemployment benefits by 19.5% and temporarily exacerbated poverty and income inequality. However, institutional factors are far more important for understanding the underlying trend, in our view. First, two segments of the Israeli society — Arabs and ultra-Orthodox Jews — constitute 60% of the poor. This is due to cultural preferences and inadequate educational attainments that limit the participation of these groups in the labor force (see Poverty by Choice, July 12, 2006). Second, globalization and the rise of Israel’s technology-intensive sectors have resulted in a dual economy, in which higher demand for skilled workers relative to the demand for less-skilled labor input leads to a wider earnings gap and thereby a worsening in relative poverty indicators (see The Achilles’ Heel of the High-Tech Economy, September 22, 2006). Of course, as Israel’s own experience with welfare spending has shown, transfer payments to households cannot eradicate poverty on a sustainable basis.

Increasing employment, not welfare spending, will deal with poverty. International figures show that the poverty risk is predominantly a function of educational attainments, which become even more important in Israel’s high-technology economy. Indeed, the incidence of poverty increases from 11.2% among Israelis with 16 years of formal schooling to 47.8% for those with less than eight years of schooling. This is a direct result of the level of participation in the labor market, which increases from 23.5% among Israelis with less than eight years of education to 77.3% among Israelis with at least 16 years of schooling. Therefore, policies aiming to increase employment, not unsustainable welfare spending, are the key to eradicating poverty and improving income distribution.

South Africa
If it Ain’t Broke, Don’t Fix it
April 11, 2007

By Michael Kafe | Johannesburg

All so soon, the South African Reserve Bank (SARB) will hold its second Monetary Policy Committee (MPC) meeting this week. While the market is pricing in a 40% probability of a 50bp rate hike at this meeting, Morgan Stanley sticks to the view that the SARB will likely keep interest rates on hold, for a number of reasons:

First is the inflation outlook: The near-term inflation outlook has deteriorated somewhat since the last MPC meeting. For example, the local price of petrol is up 16% in the last two months, thanks to a spike in global crude prices as cold weather, buoyant demand conditions, OPEC production cuts and geopolitical tensions in the Gulf region re-exerted themselves. Coupled with this is a sharp rise in maize futures prices, a jump in government tariffs as well as a current account deficit-related bout of currency weakness that has exacerbated the already deteriorating inflation trajectory, forcing us to revise our CPIX peak from 5.7%Y to 6.1%Y, and to push out the timing of the peak from March to April/May.

Even so, we expect this possible breach of the inflation target to be very short-lived, and look for inflation to close the year around 5.3%Y, dip below 5%Y in 2Q08, before rising to 5.2% at end-2008.

The SARB’s forecasts are likely to show a similar near-term inflation trajectory, although we expect the 2008 profile to come in marginally higher than our estimate, given that Morgan Stanley still has a relatively benign 2008 oil price profile. In any case, we expect the SARB to accept the fact that it is now too late to do anything about the April/May potential breach of target, and with CPIX likely to fall back within target range in less than a quarter, we do not see the justification for an additional interest rate hike.

Second is the current account deficit:South Africa’s current account deficit rose to a record high of 7.8% in 4Q06. However, as the SARB rightly pointed out, the reading was distorted by a one-off jump in oil imports as local oil companies embarked on a significant stockpiling exercise that was so huge that it lifted the inventory component of the country’s Gross Domestic Expenditure to levels not seen since the peak of the gold boom in the early 1980s.

Importantly, if one strips out the excess oil imports, the deficit was no worse in 4Q than it was in 3Q. Add to this the fact that the quality of financing flows on the capital account is fast improving as the 3Q06 haemorrhage in foreign direct investment ebbs and it is difficult to make the case for a sustained external account-induced currency weakness that would warrant further policy tightening. True, the 1Q07 current account deficit will remain fairly high as a result of a R10 billion SACU transfer that was made that quarter, but the SARB knows full well that this is also a one-off event that has everything to do with intra-union trade and co-ordinated regional development, and very little to do with domestic demand excesses.

Third is inflation expectations:At the February MPC meeting, there was no inflation expectations survey. Hence the SARB had to make do with break-even inflation rates as priced by the yield spread of conventional bonds over inflation-linkers. This time, the Bureau for Economic Research did a proper survey, and while analysts’ and business people’s inflation expectations for 2007 are unlikely to have deteriorated from the 5.5% that was published in December, we think that trade unions, whose estimates have tailed the rest of the surveyed sample in recent times, are likely to nudge their 2007 estimate toward the sample average, thanks to the rising costs of food and fuel. Similarly, we will not be surprised to see analysts upgrade their 2008 estimate, although it will still likely remain below the rather high 5.6% estimated by business people and trade unions last December.

Importantly, however, although the overall average reading for both 2007 and 2008 might deteriorate marginally, we are careful to differentiate between a true deterioration in expectations versus a levelling in expectations, as discussed above. We also expect inflation expectations for 2009, which will be covered for the first time in this survey, to come in below the 2007/2008 estimates, confirming a collective faith in the SARB’s ability to maintain low and stable inflation with the current monetary policy stance.

Fourth is the M3 money supply growth:This has remained at fairly elevated levels for some time now, and rose from 22.1%Y in January to 22.9%Y in February. As we highlighted, however, the February reading was distorted by a R24 billion bond redemption, a R10 billion coupon payment and a R1.9 billion foreign bond buyback that took net credit to government to a record high. In addition, it appears that the February number included (a) provisioning for the full payment of some R6 billion SACU transfers that were originally scheduled to go through in February but were only made in March; (b) part-provisioning for a further R8.8 billion bond redemption that was made in March; as well as (c) a R6.3 billion coupon payment that was also made in March. The Treasury’s financial statements show that, after falling from R58 billion in January to R25.1 billion in February, the government’s cash balances in interest-bearing tax and loan accounts at commercial banks rose to R29.7 billion in March, lending further credence to our view that the huge payments made in March must have been pre-funded. Only time will tell, but if we are right, then subsequent M3 readings are likely to slow.

Fifth is private sector credit growth:This has also remained uncomfortably high, and in fact rose from 24.8% in January to 26.12%Y at the last count. However, it is again important to note that the February reading was driven by a jump in ‘net other loans and advances’ - a notoriously volatile category that typically encompasses distress borrowing by corporates, not households. Hence, Morgan Stanley’s view that the debate on private sector credit at this week’s MPC meeting will essentially centre on whether or not the SARB should be hiking rates to stave off growth in corporate lending that is largely supportive of the country’s much-desired capacity expansion/capital deepening cycle. Since this type of credit is not really undesirable (in fact capacity expansions are likely to free up production bottlenecks and reduce associated resource allocation costs, thereby benefiting future inflation and exports), we would expect the SARB to hold fire here.

Sixth, although headline consumption expenditure remains robust, the discretionary components have already started slowing: Latest data from the SARB Quarterly Bulletin showed that, although durable goods consumption was up from 7%Q saar in 3Q06 to 8.4%Q saar in 4Q06, thanks largely to strong recreational and entertainment spend, the key interest-sensitive components of durable goods consumption, such as furniture, household appliances and medical equipment, slowed down in 4Q06 as interest rates rose. Also, within the semi-durable goods category, non-discretionary expenditure items like household fuel and power, medical and pharmaceutical products and petroleum products showed strong growth, while the relatively more discretionary items like food and beverages, tobacco and household consumer goods were all down from their third quarter levels, suggesting that the 200bp of interest rate hikes seen last year are in fact already beginning to impact consumption patterns. We look for continued deceleration in discretionary spend going forward, and expect non-discretionary consumption spend to start slowing by the second half of this year too. Hence we do not see the need for the SARB to pull policy rates any tighter.

Seventh, secondary data are already pointing to softer economic activity this year, and any further tightening in monetary policy risks endangering the growth outlook: New vehicle sales have plummeted from 29%Y in March 2006 through 13.8%Y at the time of the first interest rate hike in June 2006 to 3%Y presently; the SACOB Business Confidence Index is down from its peak of 103.5 in December 2006 to 99.5 in March; the BER Business Confidence Index is down from 83 in 4Q06 to 81 in 1Q07; and real retail trade is down from 12%Y in November to 9.4% in January after dipping to a revised 6.7% in December.

Eighth is the currency outlook:After rallying to a high of 7.54 in anticipation of the disappointing current account deficit, US$/ZAR subsequently sold off to a low of 7.13 last week, and still looks significantly undervalued versus its emerging market peers such as the HUF, CZK and PLN, suggesting that there is room for further rand strength in the short term. Unless the SARB extrapolates the recent weakness forward and assumes that the currency will remain weak in coming months, we think that the risk of a currency-induced increase in inflationary pressures is much less now than it was a month or so ago, when it simply wasn’t clear how the market would respond to the official current account print. Morgan Stanley sticks to the view that the rand will likely close this year at no weaker than R7.50/US$.

Ninth is the continued supportive fiscal contraction that will likely give the SARB some comfort:At the time of the February MPC meeting, the expected fiscal balance for the 2006/7 fiscal year was a deficit of 0.4% of GDP, while 2007/8 was expected to come in at a meagre surplus of 0.3% of GDP. Latest information from the Treasury, however, have it that the fiscus has already reported its first ever surplus reading in the 2006/7 fiscal year (a 0.6% of GDP surplus), while the 2007/8 estimate has also been revised to what appears to be a rather conservative surplus of 0.6% of GDP too. So, could the National Treasury single-handedly fund up to half of the country’s core current account deficit without breaching the universally accepted Maastricht threshold?

Tenth is politics: Interestingly, some South Africa watchers believe that the SARB may be running out of time to implement a final rate hike ahead of the ruling ANC’s economic policy conference in June, and party presidential elections in December, by which time it would be politically less expedient to do so. We beg to differ: The SARB is a truly independent central bank that has consistently demonstrated that it has its sights firmly set on the inflation ball, and not on political sycophancy. We see no reason why that should change now.

As always, however, there are risks to our call.The most important one is oil prices. Morgan Stanley is convinced (as is the fast-disappearing contango in the futures market) that oil prices are unlikely to go much higher from here. However, the fact is that, despite an easing of the UK-Iranian stand-off last week, oil prices have not come off the boil just yet, and there is nothing stopping the SARB from again making “...a significant upward revision of the international oil price assumptions” like it did in June 2006. Were this to happen, its CPIX trajectory for 2008/2009 would come in worse than expected, and under such a scenario, the possibility of a rate hike this week would then become feasible.