下面是小编为大家整理的美国软线零售行业展望：假期依然繁荣（精选文档）,供大家参考。
Global Research
24 October 2019
US Softlines Retail UBS Evidence Lab inside: Industry Outlook: Still Bearish on Holiday
Softgood demand looks weak and the industry faces 5 additional challenges: UBS Evidence Lab"s updated US Softlines Spending Forecast shows US consumer spending intentions for apparel and accessories remain lackluster. We project a flat industry Holiday season sales growth rate (Figure 1). Plus, we continue to expect pressure on margins from: 1) elevated industry inventory levels; 2) a further shift to online shopping; 3) Tariffs; 4) an unfavorable November weather forecast (page 8); and 5) six fewer days between Thanksgiving and Christmas, y/y. UBS Evidence Lab has a proprietary US Softlines Spending Forecast: The team"s model forecasts US clothing and accessory store sales over the next 90 days. To do this, the team first creates an Apparel Spend Intention Score. This score is based on market research revealing US consumer spending intentions. The next step applies a rigorous two-phase time series modeling technique using an ARIMA process to analyze the relationship between the market research and US Census Bureau data (see p. 16). Consumers say the economy is making them apprehensive toward shopping: Market research shows 2019 is the first year since 2011 when the state of the U.S. economy is incrementally impacting consumers shopping plans (Figure 17). This year 28.2% say the economy will affect their plans vs. 27.1% last year. These consumers say they are likely to spend less and increasingly look for bargains. This information syncs with our economic statistics-based Softline sales leading indicator (Figure 5). The October leading indicator update revealed incremental weakness m/m and suggests Holiday season softgood sales will not be strong. We maintain our Cautious view on Softline stocks: Our 2H19 EPS estimates are 2% below consensus for the average stock in our coverage universe. We anticipate downward EPS revisions negatively catalyze shares. The group"s median FY1 P/E is 12x and this looks inexpensive to some. However, the group"s median 1.9x PEG ratio suggests to us Softlines stocks aren"t too cheap. We think deteriorating fundamentals likely amplify the market"s concerns around a potential recession, tariffs, and eCommerce disruption, pressuring P/Es. Our EPS estimates are most below consensus for: CRI, FL, GPS, HBI, JWN, KSS, LB, M, RL, and TIF. Two Buy- rated stocks we think can buck the trend are SKX and AEO.
Jay Sole
Analyst jay.sole@ubs.com
+1-212-713 3559
www.ubs.com/investmentresearch
This report has been prepared by UBS Securities LLC. ANALYST CERTIFICATION AND REQUIRED DISCLOSURES BEGIN ON PAGE 19. UBS does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. Retailers, Specialty Americas Equities
Table of Contents UBS Evidence Lab Softlines Spending Forecast ....................... 3 Economic Indicators ........................................................ 6 Weather Outlook ............................................................ 8 Appendix A – Industry Statistics ...................................... 9
Appendix B – GAFO Outlook ..................................... 11
Appendix C – October Holiday Read Details ...................... 12
Appendix D – Softlines Valuation Table ........................... 15
Appendix E – Full Methodology ..................................... 16
Jay Sole
Analyst jay.sole@ubs.com
+1-212-713 3559
UBS Evidence Lab Softlines Spending Forecast
Figure 1: The UBS Evidence Lab Softlines Spending Forecast predicts flattish Holiday season growth. 10%
8%
6%
4%
2%
0%
-2%
-4%
Source: US Census, UBS Evidence Lab Inside this section: The UBS Evidence Lab Softlines Spending Forecast predicts low positive growth over the next 90 days: This includes no change y/y in November, a -0.1% decline in December, and a 0.7% increase in January. The root mean squared error out-of-sample (average from model output 4 periods ahead) is 0.75%: The model output standard deviation is 1.1% one month out and increases to 2.0% three months out. The "upper band" and "lower band" in the chart represent this range of possible outcomes. Fashion, weather, and current events are examples of factors which can cause actual results to deviate from the forecast. See the backtest in Figure 4 and methodology summary below that. Actual values Model output out-of-sample Lower band Fitted values in-sample Model projection Upper band Clothing and Accessory - Monthly Sales - 3Mo rolling Y/Y growth 01/01/2014 01/04/2014 01/07/2014 01/10/2014 01/01/2015 01/04/2015 01/07/2015 01/10/2015 01/01/2016 01/04/2016 01/07/2016 01/10/2016 01/01/2017 01/04/2017 01/07/2017 01/10/2017 01/01/2018 01/04/2018 01/07/2018 01/10/2018 01/01/2019 01/04/2019 01/07/2019 01/10/2019 01/01/2020
Figure 2: UBS Evidence Lab standardized Softlines Spending Forecast by age demographic y/y growth 3.0%
2.0%
1.0%
0.0%
-1.0%
-2.0%
Source: UBS Evidence Lab
UBS Evidence Lab Spend Intention Score - 18-34 UBS Evidence Lab Spend Intention Score - 35-54 UBS Evidence Lab Spend Intention Score - 55+
Figure 3: UBS Evidence Lab standardized Softlines Spending Forecast by income demographic y/y growth 2.0%
1.0%
0.0%
-1.0%
-2.0%
Source: UBS Evidence Lab
UBS Evidence Lab Spend Intention Score - Low Income UBS Evidence Lab Spend Intention Score - Medium Income UBS Evidence Lab Spend Intention Score - High Income
Figure 4: The latest model is directionally accurate (correctly predicts an increase or decrease in growth rate of any magnitude) 69% of the time one month ahead, and is directionally accurate 71% of the time in each of the four months ahead, on average
OCTOBER model in-sample out-of-sample (1 step ahead) out-of-sample (4 step ahead) Model output Model output Model output Down Up Down Up Down Up Actual Down 62
31
18
10
102
27
Up 26
53
6
18
44
72
Sum:
172
Sum: 52
Sum: 245
Mean Directional Accuracy:
66.9%
69.2%
71.0%
Source: UBS Evidence Lab
*Methodology: The UBS Evidence Lab Data Science team created a time series model of US Census Bureau Advance Monthly Retail Trade Survey sales of Clothing and Accessory Stores in US. We build a model for the 3 month rolling YoY growth of this series. The modeling consists of a 2 step process: 1) UBS Evidence Lab Spend Intention Score – Apparel - The UBS Evidence Lab Spend Intention Score (UBS SIS) is computed using a monthly history of Market Research on consumer intentions on different categories of Apparel. Intention to spend in the next 90 days for all categories are aggregated to reflect the aggregate directional intent, and level of intensity of directionality for each month. 2) Time series model - We have introduced a more rigorous Two-phase Time Series modelling technique. In the first phase, we fit a regression model that provides a contemporaneous relationship between UBS Spend Intention Score and the Census Sales numbers – essentially weighting the evidence of consumer intention as the sole determiner of sales. The errors from this model are then fitted as an ARIMA process. This approach allowed us to see more clearly impact from changes of UBS Spend Intention Score on the final model output and also lowered RMSE (Root Mean Square Error) – measure of model errors on out-of- sample prediction. Full detailed methodology is in Appendix E with calculations is available upon request.
Economic Indicators
Figure 5: Our proprietary 90-day clothing and accessory store sales leading indicator based on US government economic data points to a lackluster Holiday season growth rate. The newest data point shows a m/m deceleration. y/y change
10%
8%
6%
4%
2%
0%
-2%
-4%
+/- 1 StDev 3-month growth of clothing and accessory stores sales Forecast 3 months ago
Correlation = 0.75
Source: UBS Research, Haver Analytics, Bloomberg
Inside this section: US clothing and accessory store sales 90-day forecast based on US government economic data: This analysis is essentially a "check" on the UBS Evidence Lab forecast. This points to negative growth in September, but positive growth in October and November, in contrast with the UBS Evidence Lab"s forecast for persistently negative sales growth. Macro dashboard: The table shows key leading economic indicators which we believe influence apparel, footwear, and accessory sales. Many of these are used in our 90-day forecast. How does this forecast differ from the UBS Evidence Lab outlook? The UBS Evidence Lab forecast is based on information received directly from consumers. That information has better predictive power of softgoods sales than US government statistics do. Many leading indicators lost their predictive power after the 2008 recession. In Appendix B, we show a GAFO forecast, which has a higher correlation with actual sales as an alternative. This indicator points to positive sales growth y/y. How do we derive this forecast?
We collect many different monthly and quarterly macro series related to consumer behavior. Quarterly series are interpolated at the monthly frequency. At any point in time, we compute the principal components of this data. For a subset of the leading principal components and their lags, we run least squares on the three- month-ahead retail sales growth, constraining the lag structure with Almon polynomials to avoid drowning in parameters. We select how many principal components to use and how many lags to apply to each. We use brute force and loop over all possibilities (up to six components and 12 lags), choosing the combination of components and lag structure that minimizes the out-of-sample three-month-ahead forecast error over the past 5 years.
Macroeconomic Dashboard Figure 6: Macroeconomic Indicator Dashboard
Current Reading
Last Reading
Sequential Change
Y/Y
Change
L-T Avg. Y/Y Chg.
L-T Max. Y/Y Chg.
L-T Min. Y/Y Chg.
Frequency
Late...
推荐访问:软线 美国 展望 美国软线零售行业展望:假期依然繁荣 美国零售业的发展趋势