MSCI - Latest Research

Extreme Risk Analysis

Tue, 27 Jul 2010 10:12:32 +0100

Risk analysis involves gaining deeper insight into the sources of risk, and evaluating whether these risks accurately reflect the views of the portfolio manager. In this paper, we show how to extend standard volatility analytics to shortfall, a measure of extreme risk. Using two examples, we show how shortfall provides a more complete and intuitive picture of risk than value at risk. In two subsequent examples we illustrate the additional perspective offered by analyzing shortfall and volatility in tandem.

Revisiting Global Small Cap

Tue, 20 Jul 2010 16:11:49 +0100

The small cap as a source of equity risk premia has been well documented in finance literature. Although small-cap stocks are often perceived as risky relative to their large-cap counterparts, they have other characteristics that may provide an opportunity for portfolio diversification and return enhancement. In particular, moving beyond the universe of large- and mid-cap stocks into the small-cap segment triples the opportunity set in terms of number of stocks available for investors. Also, there have been pronounced performance disparities between the large- and mid-cap segments and small-cap segments of equity markets. This has motivated institutional investors to seek broader exposure and to make strategic portfolio allocations to small-cap stocks, which is evident from the growth in initial funding to small cap mandates. This research insight presents the characteristics of global, small-cap stocks relative to large- and mid-cap stocks, reviews the role that global small-cap stocks can play for portfolio diversification and return enhancement and examines different investment processes to capture the small-cap premium by comparing active versus passive strategies.

Uncovering Biases within Sectors

Wed, 07 Jul 2010 11:59:10 +0100

This bulletin is part of a series highlighting the ways in which factor models can help managers with fundamental processes. Most PMs are familiar with sector trends. Sectors go through periods of under- and overperformance, and some are more risky than others. But sectors also exhibit different characteristics that can affect their performance. Bank stocks in 2007, for instance, had significant overweights in stocks with low price-to-book ratios and high leverage. These stocks were hit harder by the 2008 financial downturn than other sectors. This short piece illustrates how factor models help quantify these biases hidden within sectors.

Evaluating Stock Screens with Performance Attribution

Wed, 30 Jun 2010 15:59:55 +0100

Stock picking often is carried out with stock screens that reduce the set of eligible stocks. Such screens usually are assessed through the level and volatility of the resultant returns, mostly with associated statistics such as information ratio or Sharpe ratio. This evaluation, however, often is insufficient to uncover unintended bets, and this research bulletin suggests complementing this evaluation by conducting Barra performance attribution on the portfolios obtained from such screening strategies. For example, performance attribution indicates that the book-to-price screen performed relatively poorly in Australia because of an unintended tilt toward low-momentum stocks. In Japan, performance attribution shows that the superior performance of book-to-price over dividend yield as a selection screen from 2009 onward was largely due to non-value tilts.

The Fundamentals of Fundamental Factor Models

Wed, 30 Jun 2010 15:54:37 +0100

This paper highlights the fundamental-based origins of the factor models used at Barra. Barr Rosenberg and Vinay Marathe (1976) first discussed the theory that the effects of macroeconomic events on individual securities could be captured through microeconomic characteristics such as industry membership, financial structure, or growth orientation. This linkage between macroeconomic events and microeconomic (or fundamental) characteristics lies at the heart of the factor model. We discuss the intuition behind a fundamental factor model, showing how it is linked to traditional fundamental analysis, and point out the insights these models can provide. Our goal is to highlight the complementary role of the fundamental factor model to traditional security analysis.

The BP Oil Spill and ESG

Wed, 30 Jun 2010 15:45:16 +0100

The BP oil spill is already proving influential in environmental, social, industrial and political spheres. It will also be pivotal in institutional investment practice. There are two likely impacts: equity portfolio construction; and the concepts of ESG investing (Environmental, Social and Governance) and universal ownership.

Private and Public Real Estate - What is the Link?

Wed, 16 Jun 2010 17:15:28 +0100

In this paper, we study the relationship between private and public real estate in the US and UK markets and demonstrate a strong link between returns in each of these markets. To uncover these relationships, we correct for appraisal smoothing and properly account for the lead-lag relationship between public and private returns. We find that public returns lead private returns, even after removing appraisal smoothing in private returns. We also discuss how real estate risk changes over longer horizons. In particular, our results suggest that private and public returns become more correlated at longer investment horizons. This carries important implications for risk management and strategic asset allocation.

The BP Oil Crisis Spills Over to UK Domestic Portfolios

Wed, 16 Jun 2010 16:45:02 +0100

The oil spill in the Gulf of Mexico has turned into a catastrophic event with major environmental and financial implications. Has this stock specific event had an impact on the wider market? What are the implications for institutional portfolios? What are the long-term investment repercussions from this crisis? This research bulletin discusses these important questions facing institutional investors in the aftermath of the BP oil spill crisis.

The Curse of Olympian Spending with International Borrowing

Wed, 16 Jun 2010 16:09:17 +0100

This Research Insight examines the impact of the unfolding European sovereign debt crisis, focusing on Greece, Portugal, Ireland, Spain, and Italy (GPISI). We use the new, short-horizon Barra Integrated Model (BIM Daily) to measure sovereign bond investment risk and provide insight into this market development. First, we highlight the background of this emerging crisis, in particular the links to government debt, fiscal deficits, maturity distribution, and levels of external borrowing. Then, we show how the recent volatility in European sovereign debt markets was reflected in BIM risk forecasts and led to high risk concentrations in a European government bond portfolio. Finally, we provide an historical and qualitative perspective to evaluate the potential widening of credit contagion.

Manipulating Correlations Through Latent Drivers

Tue, 25 May 2010 14:31:35 +0100

The analysis of a possible positive relationship between economic growth and stock market returns is interesting both theoretically and practically. Investors often wonder if they should assign higher weight to countries with higher economic performance, hoping that economic growth will eventually show up in equity returns. Although this relationship seems quite intuitive, historically long-run stock price growth has fallen short of GDP growth in many countries. In this bulletin, we use long-term equity data to analyze the steps leading from GDP to stock prices, and point out several factors that could explain why GDP growth is diluted before it can reach shareholders.

Is There a Link Between GDP Growth and Equity Returns?

Tue, 18 May 2010 16:54:42 +0100

The Perils of Parity

Tue, 18 May 2010 16:30:31 +0100

This paper examines the recent trend of adding leverage to fixed income allocations of multi-asset class portfolios of large asset owners. We show that the optimality of adding leverage from a volatility-reduction perspective depends on the correlations between bonds and equities, the relative volatility of bonds versus equities, and the weights of the two asset classes in the portfolio. If correlations between bonds and equities are negative, adding leverage could reduce the volatility of a portfolio, especially if the weight in fixed income assets is low, leverage is moderate, and bonds have a low risk relative to equities. Negative correlations also increase the likelihood that adding leverage will improve the risk-return profile of the portfolio. Asset owners considering adding leverage to their fixed income allocation can examine these influences to decide whether negative correlations between bonds and equities, a low ratio of bond to equity volatility, and higher risk-adjusted returns of bonds relative to equities are likely to persist.

Momentum in Asia Pacific Stock Markets

Wed, 12 May 2010 12:18:20 +0100

This Research Bulletin considers Momentum in Asia Pacific stock markets. We find that Momentum has performed better in some markets than others, and its effectiveness varies across time. We use the new Barra Asia Pacific Equity Model to capture these variations across markets and over time. In addition, we test a Momentum timing strategy based on hedging exposure to Momentum during market crashes. The filter we apply, which is based on those months in which the Momentum factor was negative and statistically important, mitigates drawdowns from Momentum significantly at the regional level.

Sovereign Stress and Economic Growth: Scenarios for US Investors

Wed, 12 May 2010 12:14:54 +0100

This research bulletin is the first in a series covering various aspects of stress testing and scenario analysis. In this paper, we compare and contrast two historical scenarios that may be of current interest given the increasing uncertainty regarding the returns on sovereign fixed income investments. The scenarios we consider here are the 1998 Russian debt crisis and the 1994 US rate hike that followed the savings and loan (S&L) crisis. We put a stylized US pension plan through the stress tests using BarraOne, a risk platform that provides more than 60 preloaded historical scenarios from the 1970s to the present. Each scenario that we consider applies shocks to global market factors for equities, interest rates, credit spreads, FX rates, and commodities. We review the effect of the scenarios on the pension plan, and we discuss possible hedges.

Beyond Brinson: Establishing the Link Between Sector and Factor Models

Wed, 21 Apr 2010 10:58:38 +0100

Brinson sector-based attribution explains active return in terms of intuitive allocation and selection decisions. However, it cannot easily disentangle competing industry and style effects. We introduce a special type of factor model with five defining characteristics that exactly replicates the classic Brinson model. We show that this “Brinson-replicating” factor model easily extends to explain more general types of investment processes. In this extension, returns are decomposed into style effects and pure industry effects, net of styles. Moreover, much of the classic Brinson model “stock selection effect” is attributed to contributions from a handful of style factors. We show that in this framework risk and return can be attributed to the same set of decision variables. This provides a means of comparing return contributions on a risk-adjusted basis.