World Assessment. Global risks. Introduction. Chart 1: US interest rates % % 8.0. (a) Risks to the US outlook. Chart 2: US tax cut cost

Transcription

1 Global risks Introduction The latest OEF forecast shows world GDP growth halving between 2 and 21, to just 1.9% - its lowest since But this downturn is expected to be relatively short-lived, with growth recovering to over 3% in 22 and 23. In this assessment we consider the key risks to this forecast of a V - shaped world cycle, looking, in particular, at the threats of a more pronounced downturn coming from the US, Europe and Asia. (a) Risks to the US outlook Having risen at an annualised rate of almost 4½% between mid-1995 and mid-2, US growth has slowed dramatically in the last nine months, with GDP estimated to have risen by around just 1% in both 2Q4 and 21Q1, and possibly even less in Q2. So far, this downturn in growth has been driven primarily by a sharp swing in the inventory cycle, with the move to destocking cutting.2% points off annualised GDP growth in 2Q3,.6% points in 2Q4 and 3.% points in 21Q1, and by a sharp slowdown in investment growth for example, private sector investment in business equipment and software has fallen in each of the last two quarters, having risen at double-digit rates in each of the previous eight years. In contrast, while consumer spending growth has eased from over 5% growth in 1999 and 2, it was still close to 3% in 21Q1. The recently-approved tax cuts of over $7 billion, which will boost household disposable income by 1% this year (helped by a rebate of $6 per household payable in October) and so underpin consumer demand. While the cost of the tax cut falls to under $4 billion in 22, it increases to over $9 billion again in 23 and thereafter Chart 1: US interest rates % % year government bond Federal funds rate Source: Datastream Chart 2: US tax cut cost The OEF forecast sees US growth troughing in 21Q2 and a steady recovery beginning from Q3, with GDP returning close to its trend rate of 3¼- 3½% by 23. This recovery primarily reflects the impact of the recent marked relaxation in US macro policy: The sharp cuts in interest rates since the start of the year (Chart 1), with the federal funds rate down from 6.5% to 3.75% and expected to fall at least once more, and long rates down from a peak of over 6.5% in early 2 to below 5.5% now (-year benchmark yields). This is already generating a recovery in the housing sector and helping to support consumer and business confidence both directly and because of its impact on equity markets. We would stress, however, that there is still considerable uncertainty about prospects for the US, and a more prolonged downturn and, indeed, outright recession still remains a significant danger. Much of the discussion of this risk focuses on the consumer and, in particular, the implications of the fall in equity prices since early 2 (and possibility of further sharp declines). But while the bursting of the equity price bubble will 19

2 Table 1: US personal sector's holding of financial assets $ trillions Q1 (E) A Total financial assets B Corporate equities (including mutual funds and pensions) C GDP at annual rates A / C Table 2: US saving rates by income Quintile Income Top over $79,376 Second highest ,521-79,375 Middle ,1-5,52 Second lowest ,197-32, Bottom under 17,197 Total undoubtedly slow consumer demand, we do not see the consumer as the most likely trigger of a more serious downturn in the US for two reasons: Consumers are still extremely wealthy: Even after their fall over the last year, equity prices remain much higher than in the mid-199s and, as a result, so is household financial wealth. As Table 1 shows, household financial assets are estimated to have been about $26 billion in 21Q1, down from $28 billion in 2Q4 but still 2.6 times annual US GDP. This compares with a ratio of financial assets to GDP of 2.1 in 1994, which might be interpreted as a historic norm. So, US households could see the value of their equity wealth fall by a further $5 billion ie 18% - before the wealth-gdp ratio is back to its 1994 level. Saving ratios are negative only for the richer households: While the overall personal saving ratio has fallen sharply in recent years, this wholly reflects lower saving rates for the households with above-average income levels (Table 2). These households have been the main beneficiaries of the sharp rise in equity prices through the 199s. But, given their relatively high levels of income (as well as wealth), they are also relatively well-placed to absorb the impact of lower equity prices, and so may well not increase their savings sharply to try to compensate for a fall in their equity wealth. In contrast, saving rates have actually risen since 1992 for the bottom two income quintiles. And since households in these quintiles are unlikely to have much exposure to the equity market, it is unlikely that a fall in share prices would cause them to seek to save more. (It is worth noting that part of the fall in $ billion % of GDP Chart 3: US profits Profit share (RHS) Company profits ($ billion) (LHS) Source: OEF

3 Government Rest of World Household Non-financial business Table 3: US financial balances ($ billions) Sum of above the saving ratio for higher income households reflects the treatment of capital gains taxation as a deduction from income in the US personal sector accounts, even though the capital gains are not included in gross income.) Rather than the consumer, we see the US corporate sector as the most likely source of recession. In part, this reflects the collapse of the dotcom bubble. As new finance for dotcoms has dried up, so has their ability to fund lavish expansion projects, and both their investment and workforces have been slashed. But the problems for the corporate sector are much more widely spread: As growth has slowed and competition remained intense, profits have come under pressure, falling % in the year to 21Q1. Falling profits combined with still high levels of investment mean that the corporate sector s financial deficit has continued to rise, reaching $186 billion (2% of GDP) in 2. Loan standards have tightened sharply, and spreads on corporate bonds have widened markedly. Chart 4: Net percentage of domestic respondents tightening standards for C&I loans % % Small Large and medium Source: US Federal Reserve Senior Loan Officer Surveys % of GDP Chart 5: US business investment as a share of GDP Current prices Constant prices % of GDP Source: OEF Increasing doubts have been expressed about the effectiveness of some of the recent IT investment. These developments are particularly worrying given the very high level of business investment relative to GDP in the US now (Chart 5). This is particularly dramatic when measured in constant prices. But even measured in nominal terms and therefore allowing for the falling cost of computers the ratio is much higher than seen over the last 15 years. Some of this may reflect faster rates of depreciation for computers than for the more traditional parts of the capital stock, such as buildings, machine tools etc, which mean that a higher rate of gross investment is needed now simply to maintain the existing capital stock. But even allowing for this effect, after such a strong boom, there must be a real danger of a sharp correction in investment over the next year or so. Moreover, it should be stressed that a retrenchment in investment would have implications not only for the short-term economic cycle but also for longerterm growth prospects in the US. Just as booming business investment has pushed up potential output growth over the last five years, falling investment

4 Table 4: Average annual growth in US potential output 23 Growth in business investment Contribution of TFP to potential output growth Table 5: The impact of a US hard landing US Japan Eurozone Central US Central US Central US Forecast recession Forecast recession Forecast recession GDP growth 21 (%) Inflation (%) Short-term 21 interest rates would undermine it. To illustrate this sensitivity, Table 4 calculates the likely rate of US potential growth in 23 on the basis for a range of possible assumptions for business investment and growth in total factor productivity (TFP). On our central forecast of business investment growth of 5-6% a year and a contribution of TFP growth to potential growth of about.9% points a year, we estimate that US trend growth will be 3¼-3½%. But, for example, if business investment were instead to fall by 5% a year over the next three years, potential output growth would be cut to just 2.3% and even lower if the lower rate of investment were to lead to a fall in TFP growth. To illustrate the risks to the US outlook, we have constructed a variant to our central forecast that looks at the implications of an investment-led downturn. In this case, business investment falls over 5% in 22, rather than rising by 4% as in our central forecast. But the collapse in investment is also assumed to affect confidence elsewhere in the economy. For example, with trend growth likely to be adversely affected, it is assumed to undermine confidence in the equity market, which then falls by 2-25%. And consumer confidence is assumed to be adversely affected both by the fall in share prices and the shake-out of labour that will accompany any sharp retrenchment in investment. Table 5 summarises the results of this simulation. In this scenario, the US suffers an outright recession ie two quarters of negative growth in the first half of 22. Overall GDP then falls by.2% next year, rather than rising 2.7% as in our base forecast. But inflationary pressure eases, and the Fed is assumed to cut the Federal Funds rate to just 2¼% by the middle of next year. This profile for the US obviously has serious implications for the rest of the world, with European growth cut to just 1½% and Japan suffering two years of falling GDP. The focus in this assessment has been on the downside risks to the US, which are clearly the most critical for business and the financial markets in the short term. It should be noted, however, that there are upside risks to our forecast as well. In particular, with inflationary pressures still very subdued, and the worst of the inventory correction 22

5 now over, there is scope for a stronger rebound in growth in the second half of this year and early 22. And if we are understating the benefits of the new economy for productivity, investment could rise even more strongly than we foresee in the medium term. Our concern about such a scenario, though, would be its implications for the financial imbalances in the US, which would remain substantial that is, unless the new economy quickly generates substantial improvements in US competitiveness and trade performance. (b) Risks to European outlook With the US entering a pronounced downturn, and the risks to US growth firmly on the downside, many commentators were expecting Europe, and the Eurozone in particular, to jump into the driving seat of world growth this year. That is now looking less and less likely to happen. We have revised down our Eurozone forecast and, as in the US, the risks to that forecast are heavily skewed on the downside. Indeed, all the recent comment is about the risk of recession in Europe. There is no need for a recession in the Eurozone. Core inflation remains low and is expected to fall back over the next year or so, while real wage growth has been kept under control, reflecting the high levels of unemployment still prevailing in many Eurozone economies. And the major Eurozone economies have not undertaken the massive investment seen in recent years in the US, and to a smaller extent in the UK, that could mean those countries are exposed to a recessionary risk. In fact, far from needing a recession, the Eurozone badly needs a sustained period of robust growth, to bring unemployment down closer to US or UK levels. Eurozone trend growth lower than US It is worth noting here that trend growth in the Eurozone is lower than that in the US although a large part of this discrepancy is due to different ways of measuring GDP and to different population growth rates. In the Eurozone, trend GDP growth is probably around 2¼% to 2½% per year, compared to around 3¼% in the US. But the US labour force grows at around 1% a year, compared to only around ¼% in the Eurozone. And the US measures GDP, and in particular output of IT equipment, differently using chain-weights for GDP and using hedonic (quality-adjusted) price indices for IT. In the US, if a computer this year is twice as fast as its equivalent last year, it is considered to be twice the computer: ie volume production of computers has doubled, even though the number of computers produced may be unchanged. These two factors: faster labour supply growth and different ways of measuring GDP and computer prices could add as much as 1% point to US trend growth each year. In recent years, there has been much more substantial capital deepening in the US than in the Eurozone, which has led to further productivity improvements though since much of this extra capital is IT, with its faster depreciation rate, this effect is unlikely to add to US trend growth for very long. Moreover, the extra investment in IT in the US may even now be thought to boost trend growth for a period via an increase in total factor productivity due to new economy effects albeit substantially smaller than those that were hoped for at this time last year. So, overall, there are some grounds for thinking that US trend growth is fractionally higher than in the Eurozone, after controlling for labour supply growth and different measurement systems. However, comparing the headline GDP numbers, growth of 1.6% in the US this year would be equivalent to growth of around.6% in the Eurozone: very close to an outright recession. Threat of recession in the Eurozone? We do not expect in our central case that the Eurozone will suffer a recession. But recent data suggest that the risk of recession at least in Germany, France and Italy - cannot be entirely ruled out. The scale of the slowdown in our central case to 2.% GDP growth across the Eurozone this year is already a big disappointment. A recession or anything approaching a recession, in the current context, would be a disaster. How is it that the Eurozone has come to the point of flirting with disaster? US slowdown part of the explanation Part of the explanation of the deterioration in Eurozone prospects is the US slowdown: the Eurozone is exposed to the US via trade links, financial market contagion and possible spillover effects on to domestic demand via confidence. However, it is rather less exposed than, say, the UK at least via trade and equity markets. And yet, perversely, the impact on the Eurozone seems so far to be more pronounced than that on the UK. Germany in particular looks very sickly at present. Indeed, it is difficult to ascribe all of the current weakness in the major Eurozone economies to the 23

6 % Chart 6: Eurozone: Interest rates Bunds (-year yields) Euribor (3-month rates) 2.5 Jan-99 Jul-99 Jan- Jul- Jan-1 Source: Datastream % 6. impact of the US hard landing alone: Eurozone imports are falling faster than Eurozone exports, creating a positive contribution to growth from Eurozone net trade. If the US alone were to blame, then we would strongly expect a negative net trade contribution. We would also expect the Eurozone to be hit less hard than the UK: the reverse seems to be the case. and Asia is also close to recession As we will see below, the global slowdown is not restricted to the US alone: Asian economies are also weakening sharply, and risk falling into outright recession for reasons of their own. The weakness of Asia means that the impact of the US hard landing on the Eurozone is likely to be proportionately larger. But the Eurozone is a relatively closed economy, and as such should be relatively insensitive to trade shocks. Confidence effects appear to be large It is possible that confidence in the Eurozone, although relatively high before the US hard landing came about, was rather fragile perhaps reflecting the difficult years of under-performance in the runup to EMU. But the falls in business confidence in Germany starting last autumn, which are now being echoed in France, are surely excessive if all they are is a response to the US and Asian slowdown. It would appear that other factors are in play. and ECB policy is not helping The fact that thus far the Eurozone seems to be taking a bigger hit than the UK from the US hard landing may have something to do with monetary policy. In the UK, the MPC have cut rates by 75 basis points this year, starting in February, compared to only 25 basis points from the ECB in May. Eurozone interest rates have fallen less far, and remain higher than they were during the Emerging Markets crisis of 1998/99 (Chart 6), in response to what is probably a much more serious downturn. So, although monetary policy is certainly not tight by historic standards (especially given the increase in inflation that has reduced real interest rates relative to nominal), further cuts would be justified and would provide a boost to growth. But it is hard to blame monetary policy for the weakness of Eurozone growth at present. while tax cuts have been largely eroded by inflation Monetary policy is not providing the boost to growth that it could. And fiscal policy, although it has been loosened recently, is not taking up the slack: the tax cuts announced in many countries will boost nominal disposable incomes, but the recent increase in inflation due to higher oil and food prices (Chart 7) will bring real disposable incomes back to around where they were. Overall, macroeconomic policy is not acting aggressively to offset the downturn in the way that it could and perhaps should. However, while macroeconomic policy could be more helpful, it is difficult to see it as the cause of the current weakness. But these factors are not enough to explain a risk of recession Taking all these factors into account, we should look for a marked slowdown in Eurozone activity this year. The evidence so far is that we will certainly get it and then some. But very recent data indicate that the outcome could be substantially worse than that even possibly triggering an outright recession in the Eurozone. The factors above are not sufficient, even taken together, to explain why there is even a risk of recession at present. What other candidate explanations are there? % year Chart 7: Eurozone inflation Harmonised consumer prices Producer prices -3 Jan-98 Jul-98 Jan-99 Jul-99 Jan- Jul- Jan-1 Source: Datastream % year

7 German weakness One of the factors dragging the Eurozone down has been the exceptional weakness of German demand, for which there are some German-specific explanations. Germany is suffering from some of the same problems that afflicted Japan throughout the last decade. In particular, German construction investment was massively overdone after reunification in The same was true in Japan in the late 198s, where massive construction was undertaken on very optimistic expectations of the returns to that investment. When the returns turned out not to be there, investment was reined in sharply and has remained subdued throughout the last decade. German construction investment has been falling steadily as a proportion of GDP since 1995, and those falls probably have a good deal further to go. Moreover, German competitiveness has probably suffered from an overvaluation of its exchange rate within EMU: the years from the run-up to EMU have seen France and most other Eurozone economies (with the exception of Italy) outperform Germany in terms of growth consistent with an overvaluation of the DM and perhaps also the lira. In principle, this misalignment within the fixed exchange rate should be a zero-sum game: Germany s loss is another country s gain. But in practice, the psychological effect of a weak Germany is likely to have an adverse impact right across the Eurozone. An increased taste for equities has meant burnt fingers Another possible contributor to the current weakness is the increased appetite for equities that Eurozone investors have demonstrated in recent years. This means that the recent equity price falls have burnt many more fingers than previously and these are fingers unused to being burnt. Equities had performed fantastically strongly in the years up to the middle of 2, and perhaps new investors had not fully appreciated that higher expected returns means higher risk as well. If not, they have learnt that lesson by now. It is possible, for this reason, that the fall in equity prices had a greater impact on consumer demand than we would have expected. However, privately held equity wealth is still a much less important component of total wealth in the Eurozone than in the US or the UK so even if the impact of an equity price fall on demand in the Eurozone is larger than in the past, it should still not be so large as in the UK. Ageing populations and the paradox of thrift Another possible explanation of the weakness of demand in the Eurozone is the ageing of the population, with life expectancy increasing and birth rates declining. One effect of this change could be an increase in the propensity to save, as a way of securing a reasonable standard of living during the now longer period of retirement. In the short term, an increase in the proportion of income saved can actually result in a reduction in the total quantity saved: a phenomenon known as the paradox of thrift. Higher saving ratios mean lower consumption in the short run. Unless that reduction in consumption is replaced by some other component of demand (such as investment), it will lead to lower aggregate demand, lower output, lower employment and lower income. A shift to higher saving for any reason when demand is already weak can induce an economic slowdown or even a recession. It is possible that this is what we are witnessing at present. However, while this factor may have contributed to weak demand growth over the last few years, it is hard to blame it for the rapid deterioration in recent months. But the puzzle and the risk - remains Weaker Eurozone demand is to be expected this year, because of the factors outlined above, and has been factored into our forecast for some time. But the rapidity of the recent weakening is unexpected, and points to a much more worrying risk, of a totally unnecessary recession. At present it is very difficult to understand where this risk comes from in economic terms. Perhaps the Eurozone is in danger of talking itself into a recession. Certainly, its current fragility leaves it dangerously exposed to the risk of a more pronounced slowdown in the US, as in the US recession scenario above. In that scenario, Eurozone growth falls to 1.4% next year a rate that would probably involve a technical recession in Germany and Italy, and possibly also in France. (c) Asia back in recession? Similarities and differences versus 1998 Asian trends this year look remarkably similar to Nevertheless, for all the macro similarities, the underlying factors driving this latest collapse are not the same as in 1998 and this is borne out by the differences in growth patterns being seen across countries, with Taiwan the Philippines and, probably, Singapore heading for very poor years in contrast to their relatively solid performance in 25

8 1998. This also contrasts with the robust, betterthan-feared growth in Indonesia this year, supported by still firm oil prices. This emphasises that the key driver of this new slump is not individual countries lapsing into balance of payments and external debt crises (Thailand leading Korea and Indonesia down in 1997H2). Rather it is the much weaker demand in key export markets, especially in the electronics sector and general US trade, which is hurting the small, open trading economies badly - and potentially devastating those with the highest electronics exposure: Malaysia, the Philippines, Taiwan and to a lesser extent Japan, Thailand and Korea (Table 6). Q Q1 21 year Korea Malaysia Indonesia Philippines Singapore Taiwan Thailand Japan Electronics sector drives Asian exports to crisis point The once beneficial factor of Asia s strength in IT and electronics-related industries is turning into a curse in 21 the old development lesson of the risks inherent in a one product dominated economy is back with a twist this year. After a number of distortions and uncertainties in Q1, the trade and output data from the last couple of months have Malaysia Table 6: Asian Growth (annual % change) Chart 8: Export shares in GDP as % of GDP in 2 as % of GDP in 2 Singapore** Thailand Source: Datastream/OEF Taiwan Indonesia* *as % of GDP in 1999 Total exports Electronics exports Korea Philippines China* **domestic produced exports revealed just how bad Q2 will be in almost all the Asian region. (China, helped by WTO related FDI flows and fiscal packages, is yet again an exception as is India, which depends on monsoons more than Asian trade growth.) Japan back in recession The largest Asian economy, Japan, will slide into recession as already weak domestic growth is being hit by the slump in exports. Data for Q1 were poor but data for Q2 look much worse, with export volumes down even more sharply than in As GDP fell in Q1, the likely fall in Q2 implies that Japan will yet again enter a technical recession. rest of Asia close to recession too Fear of an extremely poor Q2 has been behind recent cuts in Asian growth forecasts: what looked like a V-shaped dip in Q1 is turning into a fully fledged slump in exports in Q2, putting a number of regional economies on the brink of recession. Although seasonal data may obscure these trends, the underlying performance looks likely to bring Taiwan, Singapore, the Philippines and Malaysia to at least a stand-still, if not worse. And there are only faint hopes of an improvement in Q3 - or Q4 - all depending on the robustness of US imports and a revival in the electronics and machinery sectors worldwide. Unfortunately for Asian producers, the latter revival looks much less likely than even the hoped for pick-up in overall US demand. So we cannot expect a rapid recovery in Asia. Half an Asian Crisis"? So far, we could call 21H1 halfway to an Asian crisis - the pattern being like, but not as deep as, 1998 s plunge. The central forecast assumes that trade begins to pick up slightly in H2 and that recent currency pressures ease, assisted by the ASEAN-led stability pact. However the Asian trade slump could easily be much more prolonged - even if the overall GDP performance of the US picks up. And interlinked capital outflows would put pressure on investment as well as regional currencies. Further waves of devaluations would have consequences across the board - including putting downward pressure on world manufactures prices and producers margins. Using the OEF Global Model we can estimate a scenario in which the Asian trade outlook deteriorates further in H2. Both export performance and investor sentiment could easily weaken, the latter resulting in capital outflows and losses in forex reserves. Risk premia might then rise across emerging markets, thus constraining growth on a wider basis. Currency pressure would cause policy problems in Malaysia and 26

9 China but also, as ever, Argentina. To simulate weaker Asian trade, we assume a reduction in the Asian economies estimated market growth (the country-weighted world trade variable in the OEF Model). This market growth was cut by 5-% (from 21Q2 to 22Q1) depending on the exposure to the hard-hit electronics sector. Scenario: full recession in Asia As shown in Table 7, weaker trade has adverse effects throughout Asia and, in particular, on Malaysia (especially assuming the ringgit peg holds). This is understandable, as we can see from Chart 8 that Malaysia has one of the largest exposures to electronics exports. Notably, Singapore, which has the second largest exposure to an electronics slowdown, is also hit hard. However, the predicted devaluation in the Singapore $ (around % for all the floating Asian currencies in the scenario) would allow some of the slowdown to be offset by competitiveness gains. Unsurprisingly, the adverse effects of a prolonged electronics slump are estimated as hardest in those countries that are particularly vulnerable, either through the direct impact on GDP or exchange rate effects - or both (Malaysia). If trade (and again electronics) fails to pick up in H2, we could be heading towards this scenario for emerging markets. A firmer oil price may help Indonesia and much of Latin America, but it would hinder the others. However, an oil price collapse could occur if Asia stays in recession and this would hamper the oil producers. US (and EU) growth alone is not sufficient to satisfy the growth needs of economies exposed to the electronics sector, most obviously Asia. In addition, countries such as the Philippines and Indonesia are already saddled with large budget deficits and therefore have very little room for fiscal stimulus. While this is not a problem for somewhere as financially strong Singapore, it is extremely unlikely that even these governments could offset a severe export slowdown given the magnitude of trade in their economies. The outlook for all of Asia is inevitably tied to the trade cycle and the downside risk is assessed as a sizeable 2-3% points of GDP, putting most economies in outright recession. Japan would see GDP fall this year by around 1% and the Yen would most probably stay in the range into 22. Table 7: Weaker Asian trade scenarios Effects on GDP growth in 21 and 22 US, EU, Mexico, Brazil, Eastern European emergers % point loss.25-.5% Japan, Argentina 1% China, Taiwan, Korea, Hong Kong, SE Asia Malaysia (assuming ringgit peg holds) 2-3% >3% 27