Risk. Management. Reward.

Dynamic risk parity a smart way to manage risks

If one were to summarize the potential of risk parity in a single term, diversification would probably be the most appropriate.

Understand. Act.

Risk. Management. Reward.

Risk. Management. Reward.

Content
4 Dynamic risk parity a smart way to manage risks Using an example to illustrate the benefits of risk parity Risk parity adjusts to the current market environment The advantage of volatility targets Dynamic Risk parity a necessary addition 5

7 8

10 Estimating and managing the risks 11 Longer-term benefits of the risk parity approach 12 Making the best possible use of diversification 14 Thats all very well for the past, but what about the future? 15 Understand. Act.

Imprint
Allianz Global Investors Europe GmbH Bockenheimer Landstr. 42 44 60323 Frankfurt am Main Global Capital Markets & Thematic Research Hans-Jrg Naumer (hjn) Dennis Nacken (dn) Stefan Scheurer (st)

Data origin if not otherwise noted: Thomson Reuters Datastream

3

Risk. Management. Reward.

Dynamic risk parity a smart way to manage risks

If one were to summarize the potential of risk parity in a single term, diversification would probably be the most appropriate. Over a sufficiently long period of time, diversification generates considerable additional return on average, which is why it is frequently known as a free lunch. The extent to which diversification contributes does, however, fluctuate strongly over the course of time autumn 2008 and spring 2013 are good examples of negative developments. As such, the risk parity approach should be combined with active asset allocation so as to increase returns and reduce risks.

Dr. Timo Teuber, CFA, is a Portfolio Manager in the Multi-Asset Protection Team at Allianz Global Investors. He has been managing capital preservation portfolios for institutional clients since 2011. Dr. Teuber is responsible for developing and managing the dynamic risk parity strategy that the team has been employing successfully since 2012. He is also responsible for the conceptual design of cross-asset investment models for making efficient use of risk budgets. Dr. Teuber joined Allianz Global Investors in 2009. While completing the Global Graduate Programme, he worked as a member of various teams in Germany and the USA.

Risk parity is an investment approach aimed at spreading risk evenly (parity = the state or condition of being equal). In a conventional 50 / 50 balanced portfolio, the equal balance between equities and bonds only relates to market value. However, since equities are generally exposed to greater risk than bonds, they account for a larger share of the portfolio
4

risk and dominate the risk profile and the return to a considerable extent. By contrast, the exposure associated with equities is reduced in favour of bonds in a risk parity portfolio, such that each asset class contributes one half to the portfolio risk. The reduction of overall risk is an additional side effect of this approach.

Using an example to illustrate the benefits of risk parity

For illustration purposes, a simplified yet nevertheless realistic example where volatility is the risk measure. In this simplified case, equities demonstrate 16 % volatility and bonds 4 %. For simplification purposes, the correlation between the asset classes is assumed to be zero. In this example, the volatility of the 50 / 50 balanced portfolio is 8.25 %, with equities accounting for more than 94 % of the total risk. In reality, therefore, the balanced portfolio is actually highly concentrated and largely dependent on stock market performance. By contrast, the risk parity portfolio is invested 20 % in equities and 80 % in bonds. At 4.5 %, the volatility is much lower, and the risks are diversified.

To enable fair or comparable analysis of the income side, the different portfolios are brought to the same risk level. Considering a bond investor who wants to accept a specific volatility (4 % in this example), the investor has several alternatives: instead of allocating 100 % to bonds, he can either invest 48.5 % in the 50 / 50 balanced portfolio, or 88.4 % in the risk parity portfolio, or 25 % in equities (see figure 2). In each case, the remaining money would be allocated to risk-free for example money market investments. To identify a starting point for the earnings side, the longterm expected return for bonds is fixed at 1.5 % p. a. and for the money market at 0.1 % p. a. The latter is fairly conservative, given the current yield curve and the historical excess returns of typical bond portfolios vis--vis the money market.

Figure 1: Market value allocation Risk allocation

50/50 Balanced Portfolio

Risk Parity

Equities 50 % Bonds 50 % Market value allocation

Equities 20 % Bonds 80 %

Bonds 0.5 % Risk allocation Equities 7.8 % Bonds 2.3 % Equities 2.3 %

The top row shows the market value allocations for the 50 / 50 balanced portfolio (left) and the risk parity portfolio (right). The bottom row shows the risk allocations for both portfolios with the height being indicative of the total risk. A classic approach first determines the market value allocation, from which the risk allocation is then derived. The risk parity (or, more generally, risk budgeting) approach determines the risk allocation, from which the market value allocation is then derived. Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

Risk. Management. Reward.

Figure 2: Various portfolios with the same volatility

Bonds Bonds 100 % Market value allocation

Risk Parity Equities 17.7 % Bonds 70.7 %

50/50 Balanced Portfolio Money Market 11.6 % Money Market 51.5 % Equities 24.3 %

Equities Money Market 75.0 % Equities 25.0 %

Bonds 24.3 % Bonds 0.2 % Equities 3.8 %

Risk allocation

Bonds 4.0 %

Bonds 2.0 % Equities 2.0 %

Equities 4.0 %

Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

Based on this assumption, the 50 / 50 balanced portfolio will only demonstrate a higher expected return than the risk parity portfolio if the expected return on equities is more than 10 %. This means that the risk premium for equities must be at least 8.5 %, which is fairly high by historical comparison. In the event that equities demonstrate an expected return of exactly 10 %, the 25 % allocation to equities (with the remaining 75 % invested on the money market) can be expected to return just 2.58 % p. a. In spite of the extremely high risk premium for equities, the expected return for the diversified risk parity portfolio is greater, at 2.84 % p. a. As such, even if the outlook for equities looks positive, investors should still consider the diversified investment. After all, if the optimistic prospects for equities fail to materialize, a diversified investment should offer clear advantages. For some investors, the 30-year bond rally is the only reason why the risk parity strategy succeeds. In the example above, however, the earnings expectations are much higher for equities than for bonds. Nevertheless, a greater (risk-adjusted) return can be expected with the diversified risk parity allocation than from the best asset class (equities). The real reason why the strategy can be favorable is therefore not the performance contribution but the diversification effect of bonds. The potential inherent in this diversification effect is often underestimated. Which implies: the

more (independent) asset classes that are included, the more potential the approach offers.

Risk parity adjusts to the current market environment

Turbulent market phases are usually accompanied by higher risks and lower returns. By contrast, good market phases are characterised not only by higher returns but also by lower risks. If a portfolio has fixed market weights, this means that the risk allocation is prone to fluctuation. Conversely, the risk parity investment approach pro-actively adjusts the market value allocation to market conditions to remain compliant with the specified risk allocation. The above example can be continued for purposes of illustration by initially assuming a good market environment for equities with 12 % volatility, followed by bad market conditions with 20 % volatility. In the good market environment, the risk parity portfolio is 25 % invested in equities and 75 % in bonds, compared with about 16.7 % in equities and about 83.3 % in bonds in poor market conditions (see upper line in figure 3). Although the risk is evenly spread in both scenarios, the overall risk only increases moderately, from about 4.24 % to about 4.71 %. Overall volatility over both scenarios is about 4.48 %, while the expected return is around 3.27 %.

A look at the risk allocation of the fixed 20 / 80 allocation (the risk parity portfolio in moderate market conditions) shows how important adjustments are. Equities only account for 36 % of the risk in a good market environment, but more than 60 % in bad market conditions. Additionally the overall risk increases significantly from 4 % to around 5.12 %. Both the risk and return profiles deteriorate over both scenarios: Volatility increases to around 4.6 % while the return drops to approx. 3.2 %. Bearing in mind the observation that expected returns increase in a good market environment and decrease in poor market conditions, the risk-adjusted return improves, however, increasingly. The disadvantage of clinging to the fixed allocation is therefore clearly visible: The contribution to risk by the less attractive asset class rises in both scenarios; whether investors are always aware of this effect and, above all, really want it, is probably questionable. Strategic asset allocation defined as contributions to risk as is the case with risk parity therefore offers potential benefits, since exposure to less attractive asset classes is pro-actively reduced and increased to attractive asset classes.

The advantage of volatility targets

If levels of exposure are varied in the risk parity approach in order to keep volatility constant, then it is clearly recognisable in all three scenarios that investments in bonds remain constant at about 70.7 % (see bottom line in figure 3). By contrast, exposure to equities varies along with their volatility, i. e. as stock market risks rise, the allocation to equities must be reduced accordingly. As such, the asset class whose expectations change is the one that varies. Applying a volatility target to the fixed 20 / 80 allocation results equally in the exposure to equities being reduced as soon as the risk inherent in them increases, but bonds are also reduced although they are less risky relative to equities. An investment approach with a volatility target generally earns a better risk-adjusted return since the risk component clearly benefits from the constant volatility: If both scenarios good and bad market environments alternate, the actual mean volatility will likely be greater than the volatility in the

Figure 3: Various Risk Parity Allocations

Bad environment Bonds 83.3 % Equities 16.7 % Full investment Moderate environment Bonds 80 % Equities 20 % Good environment Bonds 75 % Equities 25 %

Money Market 15.1 % Investment with a 4 % target volatility Equities 14.1 % Bonds 70.7 %

Money Market 11.6 % Equities 17.7 % Bonds 70.7 %

Money Market 5.7 % Equities 23.6 % Bonds 70.7 %

Risk Parity allocation for various equity market conditions (bad environment 20 % volatility, moderate environment: 16 % volatility, good environment: 12 % volatility) and a constant bond volatility of 4 %. Upper line shows Risk Parity allocation with full investment and the bottom line shows risk parity allocation with a constant volatility of 4 %. In this case the equity investment part increases when market conditions get better and the money market investment decreases. Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013 7

Risk. Management. Reward.

moderate scenario. Investors may, however, not reap the benefits of higher returns from accepting this greater risk as very clearly illustrated by the fixed 20 / 80 allocation since the expected return is constant in all three scenarios. As such, investors are rewarded with higher risk-adjusted returns potential when volatility remains constant over the period. Added to which, risk-adjusted returns deteriorate measured by the Sharpe Ratio1 in poor market conditions purely as a result of the increased risk, even if the returns themselves remain constant, since the increasing volatility (the denominator of the equation) decreases the Sharpe Ratio. For an investment approach with a volatility target to become less attractive, however, the risk-adjusted returns would first have to rise considerably in poor market conditions, meaning that the expected returns would have to increase even more strongly. The potential advantage of a volatility target is further enhanced by the observation that expected returns tend to increase in a good market environment and decrease in poor market conditions.

earnings is not always the best course of action. A healthy amount of targeted active management is, however, advisable. As such, risk parity can be seen as an anchor portfolio for taking advantage of the diversification effect. Active opinions are then built in around this portfolio. Asset classes with positive (negative) prospects are assigned a higher (lower) risk contribution in an effort to enable promising opportunities to be exploited and unattractive risks avoided. Additionally a further indicator such as a market cycle indicator could offer an objective, transparent and understandable means of integrating active opinions relating to any asset class into the concept. The indicator could be based solely on historical prices and should include a trend and trend reversal component. The concept behind such a market cycle indicator: Capital markets frequently pursue longer-term trends that are measured by the pro-cyclical trend component. However, the markets often tend to exaggerate, which is then mapped by the anti-cyclical trend reversal component (see figure 4). Since the magnitude of the trends differs for each asset class, the value of the market cycle indicator can also differ. This exemplary method should therefore support the intelligent management of risk contributions. Greater risks are taken in those markets that promise disproportionately high expected returns. By contrast, risks are actively reduced in markets with negative prospects.

The Sharpe Ratio measures the ratio of the expected excess return, i. e. expected return minus the risk-free return, against the volatility.
1

Dynamic Risk parity a necessary addition

The example above clearly shows that the approach adopted by many investors of focusing strictly on specific asset classes in expectation of considerable additional

Figure 4: Dynamic Risk Parity approach with a market cycle indicator

Upward trend normal excessive normal Downward trend normal excessive normal

Price

Time Increase Reduce overover-weight weight Increase overweight Increase Reduce overover-weight weight Increase overweight

Price

Time

Active weight

Active weight

Partial profit taking

Partial market re-entry

Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

Macro-economic factors can also be tied in such an indicator. For example, bonds quite clearly differ strongly in terms of expected return, depending on whether interest levels are high or low, and interest rates are rising or falling. Fundamental factors, such as economic growth or inflation, obviously play a

key role in influencing return expectations for individual asset classes. As such, regimedependent2 assessments can add value (see Table 1). Based on these assessments, investors can actively deviate from this risk parity in order to ensure the best possible allocation for each and every market environment.

Statistical models with regime changes can model different states (e. g. bad, good or moderate market conditions). The model parameters can be different in each phase.

Table 1:Regime-dependant excess returns

Growth Excess Returns (p.a.) Equities Commodities Treasuries Unconditional Estimate 3.8 % 3.4 % 2.2 % Inflation

Low

Normal

High

Low

Normal

High

2.3 % 2.4 % 1.8 %

5.7 % 4.9 % 2.5 %

3.5 % 7.9 % 2.2 %

10.7 % 1.6 % 3.4 %

5.5 % 7.5 % 1.8 %

4.2 % 1.2 % 1.3 %

Regime-dependent risk premium assessment of US equities, commodities and US bonds. Since an assessment that disregards economic growth and inflation reveals clear differences affecting both the regime and asset classes, active asset allocation can optimize the generation of risk premium earnings. Historical simulation between 31/12/1970 30/12/2013, Indices used: S&P 500, S&P GSCI Total Return CME, Federal Reserve US Treasury Note Constant Maturity Not Averaged 10 Year, 3-Month Treasury Bill: Secondary Market Rate, US GDP, US Industrial Production, US Capacity Utilization, US CPI-Urban, US CPI ex Food & Energy, US PCE Price Index less Food & Energy, US PPI Finished Goods, US PPI Intermediate Goods less Foods & Feeds, US PPI Crude Materials Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013 9

Risk. Management. Reward.

Estimating and managing the risks

The above examples only used volatility as the risk measure, for purposes of simplification. However, various other risk measures can be applied when allocating risk parity portfolios since volatility is obviously not the only one. Even once investors have opted for a specific risk measure, there is no clear solution to adequately assessing the risk ex ante. Risk assessment must closely reflect the current risks with a small error term. For example, only the very latest historical data points might be used to estimate the risks. Using the most recent historical observations obviously produces delays in detecting structural fractions caused by regime changes. In this case, implied volatilities can help as forward-looking estimations of the risks facing market players. But beware: the unambiguity of the risk measure is not always given, nor is the availability of data. As such, historical simulations are neither available over the very long term, nor for a broad investment universe. Regime-dependent assessments have the advantage of recognising changes in risks early on. For example, bond risks differ depending on whether interest levels are high or low, and interest rates are rising or falling. The advantage of assessing risks on the basis of fundamental data

is that the estimates more closely resemble reality, although a larger data pool must be available to produce good assessments. Table 2 is an example illustrating the risks for US equities, US bonds and commodities, depending on economic growth and inflation. The advantage of volatility is that just a few data points are required to produce a good assessment. Thus historical simulations can stretch further back into the past and therefore be much more meaningful. The past has, however, proven that returns are anything but normally distributed: Many asset classes (especially bonds) demonstrate an asymmetric return distribution, that is moderate positive returns occur frequently and losses are usually rare, but severe. Also so-called fat tails, i. e. extreme events occur more frequently than would normally be expected. Risk management should, of course, consider precisely these non-normal attributes of returns to ensure that extreme events, such as witnessed in 2008, do not result in losses that investors are no longer able to shoulder. Added to which, investors must consider whether they only want to take account of price risks, or other risks as well, such as liquidity risks. The extreme example of 2008 shows that many asset classes were characterised by low liquidity. Sales were therefore only possible at severe discounts. Historical

Table 2: Regime-dependant risk assessment

Growth Unconditional Estimate 17.0 % 18.7 % 7.7 % Inflation

Volatility

Low

Normal

High

Low

Normal

High

Equities Commodities Treasuries

20.2 % 22.4 % 9.0 %

15.4 % 16.6 % 7.6 %

14.8 % 16.5 % 6.5 %

15.9 % 16.9 % 7.8 %

15.4 % 17.0 % 7.4 %

19.3 % 21.9 % 8.1 %

Regime-dependent risk assessment of US equities, commodities and US bonds. An assessment that disregards economic growth and inflation reveals clear differences. Since the risks differ in the individual regimes, the risk parity allocation will also differ. Historical simulation between 31/12/1970 30/12/2013, Indices used: S&P 500, S&P GSCI Total Return CME, Federal Reserve US Treasury Note Constant Maturity Not Averaged 10 Year, 3-Month Treasury Bill: Secondary Market Rate, US GDP, US Industrial Production, US Capacity Utilization, US CPI-Urban, US CPI ex Food & Energy, US PCE Price Index less Food & Energy, US PPI Finished Goods, US PPI Intermediate Goods less Foods & Feeds, US PPI Crude Materials Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013 10

simulations show, however, that even a risk management approach that assumes normal returns can produce very good risk reduction and ensure that maximum loss targets are not undercut. Investors can therefore generate an asymmetrical return profile, i. e. limit losses and let profits grow. The question of whether, and especially to what extent, risk management should be adopted, differs from one investor to the next.

Figure 5: Dynamic Risk Parity* with a historical advantage

1.0 0.8 0.6 0.4 0.2 0.0 Equities Government Commodities Dynamic Bonds Risk Parity 0.36 0.73

Sharpe Ratios (1/1970 12/2013) Historical simulation between 31/12/1970 30/12/2013, Indices used, Equities = MSCI USA Daily TR Gross USD, Government Bonds = Federal Reserve US Treasury Note Constant Maturity Not Averaged 10 Year, Commodities = S&P GSCI Total Return CME, Money Market = Treasury Bill Secondary Market 3 Month

0.33

0.28

Longer-term benefits of the risk parity approach

The benefit of a (dynamic) risk parity approach becomes obvious in a historical comparison: US equities, US treasuries and commodities have all demonstrated largely the same Sharpe ratio of about 0.3 between the beginning of 1970 and December 2013 (see figure 5). The Sharpe ratio for the dynamic risk parity strategy with market cycle indicator used for all the three asset classes is 0.73. Even the best combination of all three asset classes cannot match this result, which therefore clearly demonstrates the advantage of the dynamic risk parity concept compared to a static approach. Even breaking down the entire period by the respectively best asset class and determining the Sharpe ratio for each equity, treasury or commodity period shows that the strategy earns stable returns in all scenarios and generates at least the same average Sharpe ratio as the best asset class in each case (see figure 6). As such, investors either need very good forecasting capabilities, or they should prefer the dynamic risk parity approach.

Figure 6: Dynamic Risk Parity with a constantly high Sharpe Ratio

1.0 0.8 0.6 0.4 0.2 0.0 Equities Government Commodities Dynamic Bonds Risk Parity 1.0 0.8 0.6 0.4 0.2 0.0 0.05 Equities Government Commodities Dynamic Bonds Risk Parity 0.30 0.62 0.75 Sharpe Ratios during time periods strongly favoring US Treasuries (1/1970 12/2013) 0.27 0.12 0.70 0.71 Sharpe Ratios during time periods strongly favoring US Equities (1/1970 12/2013)

1.0 0.8 0.6 0.4 0.2 0.0 Equities Government Commodities Dynamic Risk Parity Bonds 0.07 0.13 0.62 0.72

Sharpe Ratios during time periods strongly favoring Commodities (1/1970 12/2013)

Historical simulation between 31/12/1970 30/12/2013, Indices used, MSCI Daily TR Gross USA USD, Federal Reserve US Treasury Note Constant Maturity Not Averaged 10 Year, S&P GSCI Total Return CME, Treasury Bill Secondary Market 3 Month; dynamic risk parity (*risk parity and market cycle indicator but without risk management) Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013 11

Risk. Management. Reward.

Making the best possible use of diversification

In a global and broad variant, one might start with the twelve asset classes (equities, commodities, REITs, and high-yield, emerging market, corporate, asset-backed, government and inflation-linked bonds, see figure 7). Other asset classes can, of course, be included, or originally included asset classes eliminated. The allocation procedure should, however, be intelligent to ensure the preservation of the underlying structure. The advantage of this asset class selection is not only that risk parity exists among the twelve classes, but that the parity is also maintained for various clusters which can be seen in figure 8. A look at the risk cluster, for example, shows that risk parity exists between the risk on three asset classes (equities, commodities, REITs, and high-yield and emerging market bonds) and the risk off classes (all other bonds). The allocation of the three equities (including commodities and REITs), govern-

ment bonds (incl. inflation-linked bonds) and credit (all other bonds) asset classes clusters also demonstrates risk parity. Last but not least, the allocation to the growth (equities, and high-yield and emerging market bonds), inflation (commodities, REITs and inflation-linked bonds) and recession (all other bonds) clusters also demonstrate an adequate exposure to the business cycle phases. In addition to the classic risk parity clusters of equities and bonds, therefore, inflation and credit are included as themes with an aim to ensure strategically good positioning looking forward. Due to the availability of data, such a broad portfolio can unfortunately only be analysed from 2001 onwards till the end of 2013. Historical simulation produces a remarkable result: Despite including asset classes with below-average performance, even the standard risk parity approach (without active asset allocation) was in the past capable of generating an identical return with the same risk

Figure 7: Strategic Risk Allocation of the broad Dynamic Risk Parity Strategy
Growth Spread Emerging Markets Bonds High Yields Emerging Market Equities Equity Global Equities Global Reits Commodities Defensive Spreads Corporates Covered Bonds European Bonds Government

Global ex Euro Bonds Euro Inflation Bonds Global Inflation Bonds Growth & Inflation Inflation Linker

Indices used for historical simulation: Global Equities = MSCI World EUR, Emerging Markets Equities = MSCI Emerging Markets Daily EUR, Commodities = Dow Jones-UBS Commodity Index Euro Total Return, Global REITs = FTSE EPRA / NAREIT Developed Index, High Yields = Barclays Pan-European High Yield Total Return Index Value Unhedged EUR, Emerging Markets Bonds = JPMorgan EMBIG Hedged in Euro, Corporate Bonds = Barclays EuroAgg Corporate Total Return Index Value Unhedged EUR, Covered Bonds = Barclays EuroAgg Securitized Total Return Index Value Unhedged EUR, Euro Inflation Bonds = Barclays Euro Govt Inflation-Linked Bond Index All Maturities, Global Inflation Bonds = Barclays World Government Ex-Euro Inflation-Linked Bond Index Hedged EUR, Global ex Euro Bonds = Barclays Global Aggregate Ex Treasury (EUR) Total Return Index Value Hedged EUR, European Bonds = Barclays Euro Agg Treasury Aaa Total Return Index Value Unhedged EUR, Global Bonds = Barclays Global Agg Total Return Index Value Hedged EUR, Money Market = Euribor 1 Month ACT / 360 Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

12

exposure as government bonds, which were the best asset class in the period in question. Active asset allocation using a market cycle indicator was even able to raise the average return by about 65 base points p. a. without increasing the risk. Furthermore, the realised Figure 8: Risk Parity in all dimensions
Risk

annual losses could be limited to less than 5 % with the aid of risk management while at the same time the average return increased due to the severe slump in 2008 (see first three lines of Table 1).

Fundamental

Asset classes Credit Growth Spread

Investment themes Defensive Spreads

Global Growth

Global Defensive

Nominal Growth

Recession Equity Government

Risk on

Risk off Inflation

Equity

Government Growth & Inflation Inflation Linker

Risk parity from all perspectives: Parity-driven breakdown of the asset classes in Risk on and Risk off clusters. Paritydriven assignment of asset classes by sensitivity for various economic phases (nominal growth, recession and inflation). Parity-driven breakdown of the asset classes credit, equities and government bonds. The resulting investment themes Growth spread, Defensive spread, Equities, Government bonds, Growth & Inflation and Inflation linker are all based on parity considerations. Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

Table 3: Results of historical simulation

Historical Simulation 1 / 2001 12 / 2013 Return (p. a.) Volatility (p. a.) Sharpe Ratio Dynamic Risk Parity (VaR 95 %, 1 year: 5 %) 5.88 % 2.51 % 1.44

Global Bonds

Risk Parity 1.0*

Dynamic Risk Parity **

Money Market (1 m EUR)

4.64 % 2.55 % 0.93

4.89 % 2.67 % 0.98

5.56 % 2.81 % 1.17

2.27 %

Results of historical simulation from January 2001 to end December 2013 for global bonds, risk parity 1.0* (*with volatility target on government bond volatility); dynamic risk parity ** (**with volatility target on government bond volatility and market cycle indicator but without risk management) and dynamic risk parity (with risk management: maximum loss target of 5 % over the course of one year). Past performance is not a reliable indicator of future results. Source: Allianz Global Investors Capital Markets & Thematic Research, December 2013

13

Risk. Management. Reward.

Thats all very well for the past, but what about the future?
Expectations can be defined for the different asset classes of the dynamic risk parity strategy based on expectations for the single assets returns and risks, which can be derived from current market data using scenario anal-

ysis and Monte Carlo simulation. Generally speaking, returns potential cannot be raised without increasing risk. Based on the current allocation, the dynamic risk parity strategy demonstrates relatively attractive risk / return profile compared to all other asset classes (see figure 9).

Figure 9: Expected Risk-Return Profile of Dynamic Risk Parity in comparison

8.0 % 7.0 % Expected Return (p. a.) 6.0 % 5.0 % 4.0 % 3.0 % 2.0 % 1.0 % 0.0 % 0% Equities World High Yield Emerging Market bonds Real Estate Equities Emerging Markets

Commodities

Dynamic Risk-Parity Corporate Bonds Covered Bonds Pfandbriefe Inflation linked bonds Government bonds US Government bonds EUR Money Market (1m EUR) 5% 10 %

15 %

20 %

25 %

Expected Volatility (p. a.)

Expected returns and volatilities for various asset classes and 1-month Euribor money market rates. Expectations for the dynamic risk parity strategy based on an asset allocation example. Past performance is not a reliable indicator of future results. Source: Risk lab Allianz Global Investors Capital Markets & Thematic Research, December 2013

14

Understand. Act.
There are various reasons why the risk parity strategy appears to have potential. The key driver of success is, however, clearly the diversification effect that enables a good ratio of return to risk by intelligently allocating risk. Since risk parity adjusts to current market conditions, timely assessment of the risks and appropriate adjustments are crucial. Linking the investment approach to volatility targets can further enhance the possibility of additional earnings. That said, active manage-

ment is enormously important, even for this investment approach, to enable conscious deviations from the anchor portfolio so as to enhance returns and reduce risks. Very longterm historical simulations indicated that the risk parity approach appeared to be superior to static allocations, even over a complete interest rate cycle. A very broad investment universe can, moreover, take even better advantage of the diversification effect. Timo Teuber

15

Risk. Management. Reward.

Do you know the other publications of Allianz GI Global Capital Markets & Thematic Research
Risk. Management. Reward. Smart Risk investing in times of financial repression Strategic Asset Allocation Managing Risk in a time of Deleveraging Active Management The New Zoology of Investment Risk Management Constant Proportion Portfolio Insurance (CPPI) Portfolio Health Check: Preparing for Financial Repression Financial Repression Shrinking mountains of debt International monetary policy in the era of financial repression: a paradigm shift Financial Repression and Regulation: Paradigm Shift for Insurance Companies & Institutions for Occupational Retirement Provision Silent Deleveraging or debt haircut? that is the question Financial Repression A silent way to reduce debt Financial Repression It is happening already EMU You can find our wide-ranging supply of publications on the euro on our Market Insights section Bonds Duration Risk: Anatomy of modern bond bear markets Emerging Market currencies are likely to appreciate in the coming years High Yield corporate bonds US High-Yield Bond Market Large, Liquid, Attractive Credit Spread Compensation for Default Corporate Bonds Why Asian Bonds? Behavioral Finance Reining in Lack of Investor Discipline: The Ulysses Strategy Overcoming Investor Paralysis: Invest more tomorrow Outsmart yourself! Investors are only human too Two minds at work Dividends Dividend Strategies and Troughs in Earnings Revisions Dividend Stocks an attractive addition to a portfolio Dividend strategies in an environment of inflation and deflation High payout ratio = high earnings growth in the future Changing World Renewable Energies Investing against the climate change The green Kondratieff Crises: The Creative Power of Destruction Demography Pension Discount rates low on the reporting dates IFRS Accounting of Pension Obligations Demographic Turning Point (Part 1) Pension Systems in a Demographic Transition (Part 2) Demography as an Investment Opportunity (Part 3)

All our publications, analysis and studies can be found on the following webpage: http://www.allianzglobalinvestors.com
@AllianzGI_VIEW

www.twitter.com/AllianzGI_VIEW

16

Investing involves risk. The value of an investment and the income from it will fluctuate and investors may not get back the principal invested. Past performance is not indicative of future performance. This is a marketing communication. It is for informational purposes only. This document does not constitute investment advice or a recommendation to buy, sell or hold any security and shall not be deemed an offer to sell or a solicitation of an offer to buy any security. The hypothetical performance and simulations shown are for illustrative purposes only and do not represent actual performance; they are not a reliable indicator for future results. Back-testings and hypothetical or simulated performance data has many inherent limitations only some of which are described as follows: (i) It is designed with the benefit of hindsight, based on historical data, and does not reflect the impact that certain economic and market factors might have had on the decision-making process, if a clients portfolio had actually been managed. No back-testings, hypothetical or simulated performance can completely account for the impact of financial risk in actual performance. (ii) It does not reflect actual transactions and cannot accurately account for the ability to withstand losses. (iii) The information is based, in part, on hypothetical assumptions made for modeling purposes that may not be realized in the actual management of portfolios. No representation or warranty is made as to the reasonableness of the assumptions made or that all assumptions used in achieving the returns have been stated or fully considered. Assumption changes may have a material impact on the model returns presented. The back-testing of performance differs from actual portfolio performance because the investment strategy may be adjusted at any time, for any reason. Investors should not assume that they will experience a performance similar to the back-testings, hypothetical or simulated performance shown. Material differences between back-testings, hypothetical or simulated performance results and actual results subsequently achieved by any investment strategy are possible. The views and opinions expressed herein, which are subject to change without notice, are those of the issuer or its affiliated companies at the time of publication. Certain data used are derived from various sources believed to be reliable, but the accuracy or completeness of the data is not guaranteed and no liability is assumed for any direct or consequential losses arising from their use. The duplication, publication, extraction or transmission of the contents, irrespective of the form, is not permitted. This material has not been reviewed by any regulatory authorities. In mainland China, it is used only as supporting material to the offshore investment products offered by commercial banks under the Qualified Domestic Institutional Investors scheme pursuant to applicable rules and regulations. This document is being distributed by the following Allianz Global Investors companies: Allianz Global Investors U.S. LLC, an investment adviser registered with the U.S. Securities and Exchange Commission (SEC); Allianz Global Investors Europe GmbH, an investment company in Germany, authorized by the German Bundesanstalt fr Finanzdienstleistungsaufsicht (BaFin); Allianz Global Investors Hong Kong Ltd. and RCM Asia Pacific Ltd., licensed by the Hong Kong Securities and Futures Commission; Allianz Global Investors Singapore Ltd., regulated by the Monetary Authority of Singapore [Company Registration No. 199907169Z]; and Allianz Global Investors Japan Co., Ltd., registered in Japan as a Financial Instruments Business Operator; Allianz Global Investors Korea Ltd., licensed by the Korea Financial Services Commission; and Allianz Global Investors Taiwan Ltd., licensed by Financial Supervisory Commission in Taiwan.

17

www.allianzglobalinvestors.com Allianz Global Investors Europe GmbH Bockenheimer Landstr. 42 44 60323 Frankfurt am Main January 2014