Published on

  • Be the first to comment

  • Be the first to like this


  1. 1. Investment research - general market conditions Research July 30, 2008 Peter Possing Andersen, Senior Analyst, +45 33 44 20 67, US: Taking stock of the bear market • Over the past year, equity markets have been declining as problems in the financial sector have fu- elled risk aversion and the earnings outlook has deteriorated. In the US, the equity market is now flirt- ing with bear market territory. • The plunge in equity wealth – in the US as well as globally – adds another headwind to the economy on top of the housing market correction, tighter credit, and rising commodity prices. However, for a num- ber of reasons the economy is currently less sensitive to equity market swings than during the previ- ous downturn in 2000-2003. Firstly, household equity holdings now account for a smaller share of household balance sheets and are smaller relative to disposable income. Secondly, corporate imbal- ances are smaller this time around. • During this expansion, equity markets have been much less important for GDP growth than in the pre- vious expansion. Over the past four quarters, the stock market has turned from adding a modest ½pp to GDP growth to subtracting a modest ¼pp from growth. Unless equity markets weaken significantly further, this headwind is not set to intensify much. Another squeeze on the economy Flirting with bear market territory 110 Index Index 110 NASDAQ (peak 31/10/07) S&P500 (peak 09/10/07) The bear is back Dow Jones (peak 19/07/07) 100 100 Over the past few quarters, the US economy has come under increasing pressure as a slowing la- percent percent 90 90 bour market and high inflation take their toll on household real income growth. Further, declining 80 80 housing wealth and tighter credit conditions are Bear market territory making it more difficult for consumers to smooth 70 70 their way through the current income squeeze. Jan Mar May Jul Sep Nov Jan Mar May Jul 07 08 . Note: Bear market territory is usually defined as a 20% de- On top of this malaise, equity markets have nose- cline in a stock index from its peak. dived recently. Two of the three major US stock market indices are now flirting with bear market Below, we take a closer look at how equity markets territory, ie, losses running close to 20%. This is impact on consumer spending and corporate in- adding to the already brisk headwinds facing the US vestment behaviour. economy
  2. 2. RESEARCH JULY 30, 2008 Household wealth under pressure Equity wealth-to-income ratio is declining Generally, the impact from equity markets on con- 200 % of nom. disp. income % of nom. disp. income 200 sumer spending growth is governed by four factors: 175 175 Equity wealth 150 150 1) Size of household equity holdings 125 125 percent percent 2) Propensity to consume out of equity wealth 100 100 3) Size of the decline in equity prices 75 75 4) Duration of the decline in equity prices 50 50 Q2-Q3 estimate (S&P500 unch. at curr. level) 25 25 In Q1, direct equity holdings (ie, outside pension 55 60 65 70 75 80 85 90 95 00 05 funds) accounted for around 17% of total house- . Note: Directly held equity wealth is defined as the sum of hold financial and housing assets. Hence, only a mi- corporate equities and mutual fund shares on household nor share of wealth effects is driven by equities. balance sheets. Even so, changes in equity prices may have a Importantly, the size of household equity holdings measurable impact on consumer spending. This is relative to disposable income is much smaller to- particularly so because equity prices tend to move day than it was in 2000 (see graph above), when faster and further than other asset prices. the last bear market occurred. Roughly speaking, this implies that, all else equal, a given decline in Household financial and housing assets the stock market will have less than half the impact $34.43 tn 54% on consumer spending today compared to 2001. $9.63 tn According to our US consumption model Research 17% US: Consumers under siege February 2008 and the appendix) the long-term propensity to consume out of financial wealth is 5 cents per dollar – an es- timate which is well within the range of estimates in the economic literature. $19.71 tn 31% The chart below shows the estimated impact on consumer spending growth of equity wealth until Financial assets ex. equity Equities Q3 this year, assuming an average S&P500 read- Housing assets ing of 1250 for the current quarter. Since late The relevant measure to apply with respect to the 2007, equity markets have turned from contribut- impact on consumer spending growth is equity ing ½pp to consumer spending growth to having a wealth relative to nominal disposable income. This broadly neutral effect on consumer spending. ratio declined from 109% in Q3 2007 to 91.7% in Q1 2008. With S&P500 hovering around 1250, Estimated wealth effect from equities on consumer the ratio is likely to dip to around 80% in Q3 this spending growth 1960-present … 2.0 %-point q/q AR %-point q/q AR 2.0 year. 1.5 1.5 1.0 1.0 0.5 0.5 percent percent 0.0 0.0 -0.5 -0.5 -1.0 -1.0 Wealth effect from equity prices -1.5 on consumer spending -1.5 60 65 70 75 80 85 90 95 00 05 . DANSKE BANK 2
  3. 3. RESEARCH JULY 30, 2008 … 2000 - present … and their impact on consumption growth 2.0 %-point q/q AR %-point q/q AR 2.0 2.0 %-point q/q AR %-point q/q AR 2.0 1.5 1.5 1.5 1.5 Wealth effect from equity prices Wealth effect from equity prices 1.0 on consumer spending 1.0 1.0 on consumer spending 1.0 0.5 0.5 0.5 1) 0.5 percent percent percent percent 0.0 0.0 0.0 0.0 -0.5 -0.5 -0.5 2) -0.5 3) -1.0 -1.0 -1.0 -1.0 -1.5 -1.5 -1.5 -1.5 00 01 02 03 04 05 06 07 08 00 01 02 03 04 05 06 07 08 09 . . While this effect does not seem very dramatic, we The simulations provided several relevant observa- have not yet seen the full effect of the current de- tions. cline in equity prices, as it will take some time to feed fully through. It is important to realise that, Firstly, in all three scenarios the negative im- generally, wealth effects do not impact on a day-to- pact from equity prices will increase in H2 – day basis, but gradually over several quarters. once again due to the lagging nature of wealth Hence, not only the magnitude, but also the dura- effects. tion, of equity price cycles is an important factor in the overall impact on consumer spending. Secondly, none of the scenarios reveal their full effect on consumption until next year, as the im- In order to fully judge the equity wealth impact, we pact of lower equity prices takes time to feed need to have knowledge of not only how deep the through. According to our scenarios, equity equity market correction is set to be, but also of prices will impact annual consumer spending how long time it will take to play out. growth by between -½pp and +¼pp. To get an idea of the dynamics, we have drawn up a Thirdly, consumers are less sensitive to equity couple of scenarios based on the S&P500 index. markets today compared to the previous bear market in 2000-2002 (see discussion above). 1) S&P500 recovers to 1500 by year-end Even in our most negative scenario, which re- and then rises by 10% in 2009 sembles a decline in the stock market similar to 2) S&P500 flat at 1250 until end-09 that in 2000-2002, the drag on consumption 3) S&P500 declines to 1000 by year-end and will only marginally exceed ½pp. Back then, de- the stays flat in 2009 clining equity prices took out around 1pp of consumption growth. Three stock market scenarios… 1700 Index Index 1700 Implications for corporate spending 1600 1600 1) It is not only consumption that is sensitive to equity 1500 1500 1400 S&P500, monthly avg. 1400 markets, investment is too. However, the invest- 1300 1300 ment channel is more subtle, as the impact can percent percent 1200 2) 1200 come directly from equity valuation, or indirectly, as 1100 1100 an effect of future expected earnings – ie, changes 1000 1000 in demand. 3) 900 900 800 800 00 01 02 03 04 05 06 07 08 09 As the chart below illustrates, the relationship be- . tween changes in equity prices and corporate spending growth does indeed exist, but it is rela- tively unstable over time. DANSKE BANK 3
  4. 4. RESEARCH JULY 30, 2008 S&P500 vs. non-residential investment spending lation and accelerator impact) on corporate spend- 45 % y/y << S&P500 deflated* %y/y 23 ing from changes in equity prices can be measured 35 18 from coefficients in the regression above (see chart 25 13 below). 15 8 5 percent percent -5 3 Impact on corporate spending from equity market -3 6 %-point q/q AR %-point q/q AR 6 -15 -25 -8 4 4 -35 Corporate spending, -13 constant prices >> 2 2 -45 -18 1) 50 55 60 65 70 75 80 85 90 95 00 05 0 0 percent percent 2) . -2 -2 Note: SP500 has been deflated with the national accounts deflator for non-residential investments. -4 -4 3) Effect on corporate -6 spending from equity prices -6 Correlations on a quarterly basis generally suggest -8 (in deviation from trend) -8 that changes in equity markets lead corporate 96 98 00 02 04 06 08 spending by approximately two quarters. . Note: Impact is measured as deviation from trend, and in- cludes the direct impact from equities, the impact via con- sumer spending, autocorrelation and the GDP accelerator S&P500 vs. corporate investment spending effect. in the three scenarios 45 % y/y << S&P500 deflated* %y/y 23 The results suggest that equity markets are al- lagged 2 qtr's 35 18 ready dampening corporate spending moderately 25 13 15 1) (relatively to trend). Moreover, the drag is likely to 8 5 intensify somewhat if equity markets do not re- percent percent 3 -5 bound over the course of H2. 2) -3 -15 -25 -8 Corporate spending, constant prices >> 3) -13 Importantly, the sensitivity of corporate spending to -35 -45 -18 equities prices is likely to vary over time. In particu- 86 88 90 92 94 96 98 00 02 04 06 08 lar, corporate spending could be extra sensitive to . equity prices when there are large imbalances in However, the relationship might not be directly in- the non-financial corporate sector. However, this terpretable, as changes in equity prices mirror all does not seem to be the case this time around, as kinds of information in the market, not only what is there is little sign of overinvestment or over-hiring relevant for investment behaviour. in corporate businesses. This is an important ar- gument for corporate spending being more resilient To quantify the impact of lower equity prices on this time. corporate spending we estimated a model based on current and lagged consumer spending, exports, Adding it all up S&P500 and lagged changes in annual GDP growth By summing the impact on consumer and corpo- (an accelerator component). This model is pre- rate spending we end up with the total impact on sented in the chart below (see specifications in ap- GDP growth. pendix). Total impact on GDP 1960-present… Modelling corporate spending 2.0 %-point q/q AR % y/y 45 45 % q/q AR % q/q AR 45 S&P500 >> 1.5 35 Corporate spending, 35 accellerator model 35 1.0 25 25 25 1) 0.5 15 15 15 percent percent 0.0 5 2) percent percent 5 5 -0.5 -5 -5 -5 -1.0 -15 3) -15 -15 << Estimated GDP -1.5 -25 Corporate spending impact from equity prices -25 -25 -2.0 -35 -35 -35 60 65 70 75 80 85 90 95 00 05 10 50 55 60 65 70 75 80 85 90 95 00 05 . . The impacts (a direct impact from equities and in- direct impacts via consumption impact, autocorre- DANSKE BANK 4
  5. 5. RESEARCH JULY 30, 2008 …2000 to present to GDP growth compared to 1-1½pp in the late 2.0 %-point q/q AR % y/y 45 1990s. 1.5 S&P500 >> 35 1.0 25 The effect of equity prices on GDP growth has 1) 0.5 15 gradually turned negative in 2008: equity markets percent percent 0.0 5 2) are currently subtracting around ¼pp from GDP -0.5 -5 growth on an annualised basis. -1.0 -15 3) -1.5 << Estimated GDP -25 impact from equity prices Unless equity markets fall another 20%, this effect -2.0 -35 00 01 02 03 04 05 06 07 08 09 is unlikely to become substantially more negative . over the coming quarters. On the other hand, if eq- From our estimations it is evident that equity mar- uity markets rebound in H2, they could boost the kets have been far less important for the current economy mildly in 2009. expansion than was the case for the expansion in the late 1990s. In the current expansion, rising eq- uity prices have been adding between ¼pp and ½pp DANSKE BANK 5
  6. 6. RESEARCH JULY 30, 2008 Appendix: Consumer model The model for consumer spending which has been applied throughout this piece of research is an error-correction model. Similar models are used by the Federal Reserve staff (see Morris Davis and Michael Palumbo, “A primer on the Economics and Time Series Econometrics of Wealth Effects.” Federal Reserve FEDS paper 2001-9). The model is a two-step error correction model consisting of a long-run and a short-run equation. The estimation period is 1960-present. The long-run equation determines the ratio between the desired long-term target for outlays (CTARGET) and incomes (Y). This equa- tion is based on the permanent income hypothesis that households perfectly smooth consumption over time in accordance with asset wealth and human wealth (or total future income). The equation is specified using separate propensities to consume out of net housing wealth (NWHOUSE) and net financial wealth (NWFIN ). Further transfer income (TR) is included to capture structural changes in the population, ie, changes in the active labour force relative to the total population. The idea is that the person who is not in the labour market (unemployed, young people in education, pensioners) tends to save less. A short-run equation is modelled around the long-run equation. This relates changes in the consumption ratio (ΔC/Y) to • Lagged change in the consumption ratio, ΔC/Y • The deviation from the target spending , {CTARGET/Y – C/Y } • A measure of the change in credit growth, CREDIT • Changes in the unemployment rate, ΔU • The real fed funds interest rate, RFF. This is specified as nominal fed funds minus long-run inflation expectations from Survey of Professional Forecasters • Nominal disposable income growth, Y(t)/ Y(t-1) • Headline PCE inflation, π Long-run equation: CTARGET/Y = 0.65 + 0.05*NWFIN/Y + 0.07*NWHOUSE/Y+ 0.35*TR/Y Short-run equation: Δ{C/Y}(t) = 0.86 – 0.13*Δ{C/Y}(t-1) - 0.15*{CTARGET/Y(t-1) – C/Y(t-1)} + 1.08*CREDIT(t) – 0.41*ΔU(t) – 0.05* RFF(t) – 0.68*{Y(t)/ Y(t-1)} + 0.51*π(t) CREDIT(t) = Δ{ΔConsumerCredit/Y}(t) + 0.25* [ {NetMEWavg4/Y}(t)-{NetMEWavg4/Y}(t-4) ] /4 NetMEWavg4(t)=[ {NetMEWavg4/Y}(t)+ {NetMEWavg4/Y}(t-1)+ {NetMEWavg4/Y}(t-2)+ {NetMEWavg4/Y}(t-3) ] /4 Appendix: Corporate spending model The model for corporate spending, CS(t), is estimated as an autoregressive ordinary least squares regression. The estimation pe- riod is 1950 to present. The following explanatory variable is included • Annual acceleration in annual GDP growth, ΔΔGDP • Quarterly personal spending growth, ΔC • Quarterly export growth, ΔE • Quarterly growth in S&P500, ΔSP500 Model equation: ΔCS(t) = -4.35 + 0.21*ΔCS(t-1) + 0.34* ΔΔGDP(t-2) + 1.45*ΔC(t) + 0.44*Δ(t-1) +0.09*ΔE(t) + 0.07*ΔE(t-1) + 0.05*ΔSP500 DANSKE BANK 6
  7. 7. RESEARCH JULY 30, 2008 This report has been prepared by Danske Research, which is part of Danske Markets, a division of Danske Bank. Danske Bank is un- der supervision by the Danish Financial Supervisory Authority. Danske Bank has established procedures to prevent conflicts of interest and to ensure the provision of high quality research based on research objectivity and independence. These procedures are documented in the Danske Bank Research Policy. Employees within the Danske Bank Research Departments have been instructed that any request that might impair the objectivity and independence of re- search shall be referred to Research Management and to the Compliance Officer. Danske Bank Research departments are organised independently from and do not report to other Danske Bank business areas. Research analysts are remunerated in part based on the over-all profitability of Danske Bank, which includes investment banking revenues, but do not receive bonuses or other remuneration linked to specific corporate finance or dept capital transactions. Danske Bank research reports are prepared in accordance with the Danish Society of Investment Professionals’ Ethical rules and the Recommendations of the Danish and Norwegian Securities Dealers Associations. Calculations and presentations in this report are based on standard econometric tools and methodology. Documentation can be ob- tained from the above named authors upon request. Major risks connected with recommendations or opinions in this report, including as sensitivity analysis of relevant assumptions, are stated throughout the text. First date of publication Please see the front page of this research report. This publication has been prepared by Danske Bank for information purposes only. It is not an offer or solicitation of any offer to pur- chase or sell any financial instrument. Whilst reasonable care has been taken to ensure that its contents are not untrue or mislead- ing, no representation is made as to its accuracy or completeness and no liability is accepted for any loss arising from reliance on it. Danske Bank, its affiliates or staff, may perform services for, solicit business from, hold long or short positions in, or otherwise be in- terested in the investments (including derivatives), of any issuer mentioned herein. Danske Bank’s research analysts are not permit- ted to invest in securities under coverage in their research sector. This publication is not intended for private customers in the UK or any person in the US. Danske Bank A/S is regulated by the FSA for the conduct of designated investment business in the UK and is a member of the London Stock Exchange. Copyright (©) Danske Bank A/S. All rights reserved. This publication is protected by copyright and may not be reproduced in whole or in part without permission. DANSKE BANK 7