Forex Weekly Reports Economics Weekly: Risk of Recession Grows
| Sunday, 07 September 2008 16:50 | |||||
|
| Economics WeeklyRisk of Recession GrowsSterling is unlikely to find support from economic data releases this week, as annual producer price growth will remain very strong but interest rates cannot be raised, BRC retail sales stay weak and the global trade deficit hover around £7.5bn. In addition, the quarterly Bank of England inflation expectations survey will almost certainly confirm a strong rise. The MPC testifies to the Treasury Select Committee, attracting interest on Thursday. In the Eurozone, the ECB publishes its monthly report, spelling out the economic forecast given by ECB President Trichet at last week's press conference, while his testament to the European Parliament's Committee on Wednesday will keep markets alert. Whilst digesting the implications of Friday's very weak US monthly labour report, focus this week turns to the still large US trade deficit, August retail sales growth and the September University of Michigan consumer confidence survey. Following the RBA's decision to cut Australian interest rates by 0.25% to 7% last week, New Zealand's central bank is likely to follow suit and cut rates by 0.25% to 7.75%, as the pace of economic growth slows. The surprise revision in UK Q2 GDP growth from +0.2% to flat and market perceptions of a more serious downside risk to H2 growth, following publication of the BoE's August inflation report and a plethora of weak business surveys and housing activity reports, have quickened the pace of sterling's decline. But along with more widespread concern about recession, near-term inflation expectations have risen. Data to inform on growth/inflation trends this week include August producer prices, which may come off peak levels, but remain extremely high by historic standards. This price data will highlight concern about the pass through to CPI inflation on the one hand, and on the other, the possibility of further margin compression in the UK corporate sector as weak market growth forces many companies to absorb higher input costs. Also due, the market consensus for industrial and manufacturing output in July is for both to contract by 0.1%, but we expect a rise. On the consumer side, another weak BRC retail sales like-for-like survey may dim the retail sales outlook, while the RICS house price index may fall to -85 in August from -83.9 in July, consistent with declines in other published house price indices. Finally, the UK's global trade deficit may narrow slightly to -£7.5bn in July from -£7.7bn in June, but the deficit may improve more rapidly in the months ahead because of slower economic growth, retrenchment in commodity prices and more competitively priced exports due to the weaker pound. Focus on the US jobs market intensified following publication of a weak August labour market report. Non-farm payroll jobs fell by 84,000 and there were significant downward revisions to the previous two months' figures. Moreover, the unemployment rate rose sharply to 6.1% from 5.7% in July, the highest level in five years. The important thing here is that employment always responds with a lag to weaker GDP growth, so the risk is that job cuts become more widespread. In terms of this week's data, retail sales may have increased in August and the University of Michigan consumer confidence, also for August, may have picked up slightly from 63.0 to 63.5 (above June's low of 56.4). But it may be difficult to divert market attention away from concern over US job losses. Net exports have supported growth this year, but the US trade deficit may have widened in July to $58.0bn from $56.8bn in June, partly because oil import prices peaked in mid-July. The ECB kept its repo rate at 4.25% last week, while President Trichet re-affirmed its neutral policy bias. By focusing on the growing risk of second round price effects, the ECB is keen to dent hopes of wage growth in the wake of IG Metall, Germany's engineering trade union, indicating that it could demand wage increases of up to 8%. Data this week is limited to EU-15 industrial production, which may contract by 0.3% in July, compared with a flat outcome in June, the third consecutive monthly fall. This is related to the adverse impact of rising raw-material costs and the strong euro. However, the euro's recent depreciation may have a positive impact on Europe's manufacturing industry in the months ahead. Chart 1: UK outperforms US & Eurozone on official retail sales growth, but UK retailers' own surveys are far more downbeat
Chart 2: The weaker US$ has helped shrink the US trade deficit. Will the year-to-date 10% fall in the £ trade-weighted index do the same for the UK?
Are Economic Growth Forecasts Any Use?How accurate are economic growth forecasts and does it matter?There is a lot of attention being placed on economic growth forecasts at present, not surprisingly since growth is slowing sharply and policy makers, companies and individuals want an idea of what is going to happen next so that they can plan ahead. We have looked at forecasts of economic growth in the UK, US and Germany (as a proxy for the Eurozone) for the last decade. The results show that great care should be taken in using, and interpreting, economic growth forecasts. For instance, it is shown that the consensus forecast, which is judged more accurate than the individual forecasts that make it up, is rarely, if ever, actually spot on. In fact, for the period 1999 to 2007 (we estimate 2008), for the three countries assessed in this analysis, the consensus did not accurately predict the actual outcome once. Charts a to c show that forecasters were close some of the time but tended to be wrong all of the time. For the UK, see chart a, forecasters tended to consistently underestimate actual gdp growth (i.e. they are pessimistic), with the outturn being more often above the forecast line than not. For the US, the opposite was true, forecasters tended to keep overestimating growth (i.e. they are optimistic). Germany was a mixture of the US and the UK. Chart a: Gdp forecasts are rarely accurate: UK
Chart b: Gdp forecasts are rarely accurate: US
Chart c: Gdp forecasts are rarely accurate: Germany
This is not to say that individual forecasts were not correct (we did not look at them all), but the same forecaster rarely gets it right consistently, otherwise everyone would tend to follow that forecaster and the consensus estimate would improve. So, if, as our analysis shows, this is not the value of gdp forecasts, what is? That seems to be in getting the direction of the change in economic growth right and, by implication, the analysis of the reasons for this. Forecasts of gdp are not accurate relative to the actual outturn...Table 1 shows some summary statistics for gdp forecasts for the US, UK and Germany. Forecasts were taken from the December edition of Consensus Economics for the year ahead. The average standard forecasting error for gdp in the three economies is 1 percentage point. In other words, if the consensus gdp forecast is 2%, the average error of past forecasts (comparing the forecast in each year with the actual outcome) suggests, with an equal level of confidence, that the actual outcome could be 1% or 3%. Table 1: The (in)accuracy of gdp forecasts (1999-2008)
The forecasting error for the UK is the lowest, at 0.7 percentage points, and so differences of 0.7 percentage points or less either side of the consensus forecast are not significantly different from each other. The standard error is highest for Germany, at 1.2 percentage points either side of the consensus, and 1.1 percentage points for the US. The standard deviation, the dispersion of the actual gdp figures from its mean over the period, is close to the error of the forecasts at around 1%, showing that actual gdp growth can and does vary, on average by 1 percentage point, each year from its mean value. This gives a feel for the ‘normal’ variability to be expected in these sorts of numbers. It is highlighting that gdp is a large number and that a small change in its value can lead to big percentage variations in its growth rate from year to year. The value of gross domestic product in the US economy in 2007, for instance, was $13.8 trillion, for the UK it was $2.8 trillion and for Germany the figure was $3.3 trillion at current prices and at market exchange rates. ...but many outcomes fall within the forecast range...Table 1 also shows that in none of the countries analysed was the consensus forecast of economic growth actually spot on compared with the actual outturn in any of the years in the period. For the UK, see chart d, the actual gdp growth rate was outside of 1 standard error only twice in the last decade. It was the same for the US, chart e, but Germany’s actual outcome, chart f, was outside of its forecast range four times in the last decade, much the worst forecasting record. We think the lesson from this is not to focus too much on the actual spot gdp forecasts so much as on whether it is close to the consensus or not. And, moreover, do not focus on whether it is higher or lower than the consensus, but on the extent of the difference from the consensus. The greater the difference from the consensus, the more likely it is to be right, or wrong. In other words, a truly radical forecast is not one that is within 0.7 percentage points of the mean UK forecast for 2008 of 1.4% currently, but one that is outside of that range, i.e. less than 0.7% growth this year or more than 2.1% growth. Looking at the August 2008 consensus shows that none of the 29 forecasters are outside this range. Chart d: GDP forecasts usually get the direction of growth right: UK
Chart e: GDP forecasts usually get the direction of growth right: US
Chart f: GDP forecasts usually get the direction of growth right: Germany
But since we are into the second half of 2008, the likelihood of a truly remarkably different outcome from the consensus is significantly decreased, so how about 2009? The latest consensus forecast for 2009 is for UK gdp growth of 0.9%. How many lie outside of the 0.7 percentage points error range? The answer is 4 forecasters, or just 13.8% of the total. In other words, roughly 86% of the forecasts are clustered within the standard error for UK gdp forecasts (of the 4 dissenting forecasts, 1 is looking for an outright fall in output growth of nearly 2%, while 3 are looking for growth of 1.8-2.0%). For the US in 2009 this figure is 96% of forecasters (just 1 outside the average forecast error), for Germany the figure is 100%, with an average for the whole group of 94%. This just proves how forecasters bunch together. ...but the best use of gdp forecasts is in indicating the direction of gdp growth and the reasons for itIf forecasters are bunched together and are never actually completely accurate, what use are they? The answer lies in the direction of the forecasts. Table 1 shows that consensus forecasts get the direction of the move in economic growth correct 77% of the time, much better than a toss of the coin (for the other 23% there is no overall directional bias, but in the UK, forecasts tend to be pessimistic, for the US, optimistic and no bias for Germany. So the perfect forecast is not just one that gets the direction right but one that gets the size of the move right as well. That happens all too rarely, but at least users of economic forecasts, and economic forecasters themselves, should bear in mind some of these features of gdp forecasts when talking to users of their forecasts. This might help to avoid the trap of talking about gdp forecasts as if they are radically different when in reality they are not. Lloyds TSB Bank Disclaimer: Any documentation, reports, correspondence or other material or information in whatever form be it electronic, textual or otherwise is based on sources believed to be reliable, however neither the Bank nor its directors, officers or employees warrant accuracy, completeness or otherwise, or accept responsibility for any error, omission or other inaccuracy, or for any consequences arising from any reliance upon such information. The facts and data contained are not, and should under no circumstances be treated as an offer or solicitation to offer, to buy or sell any product, nor are they intended to be a substitute for commercial judgement or professional or legal advice, and you should not act in reliance upon any of the facts and data contained, without first obtaining professional advice relevant to your circumstances. Expressions of opinion may be subject to change without notice. Although warrants and/or derivative instruments can be utilised for the management of investment risk, some of these products are unsuitable for many investors. The facts and data contained are therefore not intended for the use of private customers (as defined by the FSA Handbook) of Lloyds TSB Bank plc. Lloyds TSB Bank plc is authorised and regulated by the Financial Services Authority and is a signatory to the Banking Codes, and represents only the Scottish Widows and Lloyds TSB Marketing Group for life assurance, pension and investment business.             
|
||||
