Modern Deterministic Scenarios for Interest Rates Presented to the - PowerPoint PPT Presentation

Modern Deterministic Scenarios for Interest Rates Presented to the Chicago Actuarial Association March 20, 2018 Mark E. Alberts, FSA, MAAA

Agenda I. Background on Modern Deterministic Scenarios (MDS) research project II. Overview of MDS interest rate scenario development & results III. Overview of MDS analysis of common stocks, bond spreads, inflation IV. Conclusions V. Questions

I. Background on Modern Deterministic Scenarios (MDS) research project

What is (are?) Modern Deterministic Scenarios? Society of Actuaries Research report: Sponsored by Financial Reporting Section, Smaller Insurance Company Section, Committee on Life Insurance Research. Posted September 2017. Primary Research Objective: Develop a set of deterministic cash flow testing scenarios that may be considered moderately adverse in varying interest rate environments, particularly the current low rate environment Secondary Research Objectives: Provide considerations in modeling inflation, investment spreads and equity returns 4

What is (are?) Modern Deterministic Scenarios? Research Methodology:  Identify or construct historical data series to use as basis of analysis  Perform Conditional Tail Expectation (CTE) and other empirical analysis of historical data  Construct scenario algorithms based on the historical data analysis Research Output:  Report describing the research methods, assumptions and data sources  Appendices detailing the analysis  Excel workbooks to generate the MDS scenarios (Appendices J and K) https://www.soa.org/research-reports/2017/2017-modern-deterministic-scenarios/ 5

Why Modern Deterministic Scenarios? New York 7 – Most common cash flow testing scenario set  Previously required; no longer so in most states  Concerns with applicability in low interest rate environment  Other limitations – parallel shift convention; symmetry of increasing/decreasing scenarios; level of stress is untested Stochastic Scenarios – used by many, but not all, appointed actuaries  May be impractical for smaller companies  Results more difficult to interpret than deterministic scenarios  Best-estimate or best-case? 6

Why Modern Deterministic Scenarios? New York 7 – Historical prevalence of a 3% pop-up/pop-down (NY4, NY7) and +/- 5% over 10 years Short Rates Long Rates Pop-up 3% CTE98 >CTE99 Pop-down 3% CTE98 >CTE99 +5% over 10 yrs CTE96 CTE98 -5% over 10 yrs CTE93 CTE99  Higher interest rates  larger pops  For long rates, the maximum pop in either direction was 1.36% when the initial rate was 6% or less, but 3.04% when the initial rate was greater than 6%. (U.S Treasury data since 1953)  For short rates, the maximum pop in either direction was 2.33% when the initial rate was 4% or less, but 6.46% when the initial rate was greater than 4%. (U.S Treasury data since 1953)  10 year changes vary with starting rate, demonstrating mean reversion pattern 7

II. Overview of MDS interest rate scenario development & results

MDS Project Phases MDS Phase I: MDS Interest Rate Series Development MDS Phase II: Interest Rate Data Analysis MDS Phase III: MDS Scenario Construction MDS Phase IV: Common Stocks, Corporate Bond Spreads, Inflation

MDS Phase I: MDS Interest Rate Series Challenges:  Need a data series long enough to capture range of historical variation, but recent enough to be applicable  Robust yield data on US Treasuries only available back to early 1950s. Is this adequate to provide long-term expectation?  Earlier U.S. Government bond yield data generally only available aggregated by term  Earlier U.S. Government bond yields represent poor measure of market interest rates, for various reasons  The U.S. did not replace the UK as the world’s dominant economic power until after World War I , and was essentially a developing nation through the latter part of the 19 th century 10

MDS Phase I: MDS Interest Rate Series Solution:  Constructed two interest rate series – MDS Long Rate series (1729-2015) and MDS Short Rate series (1825-2015)  Treasury yields used for as long as reliably available – 90 day for the Short series, beginning 1931; average of 20 and 30 year for the Long series, beginning 1953  Short rates 1918-1930 – prime bankers acceptances, 90 day  Long rates 1923-1942 – Moody’s AAA less spread; 1943-1952 – US Government taxable bond yields, 15 years+ to call or maturity  UK government rates used through WWI 11

MDS Phase I: MDS Int Rt Series Development Interest Rates since 1953 20/30CMT Avg 16.00 14.00 12.00 10.00 8.00 6.00 4.00 2.00 - Average 1990- 1953- Source: 20/30CMT 5.39 6.21 FRB, H.15 Selected Interest Rates 12

MDS Phase I: MDS Int Rt Series Development Interest Rates since 1798 20/30CMT Avg US Long - Contemp 16 14 12 10 8 6 4 2 0 1798 1803 1808 1813 1818 1823 1828 1833 1838 1843 1848 1853 1858 1863 1868 1873 1878 1883 1888 1893 1898 1903 1908 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Average 1990- 1953- 1920- 1798+ Sources: 20/30CMT 5.39 6.21 NA NA FRB, H.15 Selected Interest Rates US Long 6.25 6.87 5.81 5.23 Measuringworth.com 13

MDS Phase I: MDS Int Rt Series Development Interest Rates since 1798 20/30CMT Avg US Long - Contemp UK Long Contemporary 16 14 12 10 8 6 4 2 0 1798 1803 1808 1813 1818 1823 1828 1833 1838 1843 1848 1853 1858 1863 1868 1873 1878 1883 1888 1893 1898 1903 1908 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Average 1990- 1953- 1920- 1798+ Sources: 20/30CMT 5.39 6.21 NA NA FRB, H.15 Selected Interest Rates US Long 6.25 6.87 5.81 5.23 Measuringworth.com UK Long 5.49 6.64 5.58 4.66 Measuringworth.com 14

MDS Phase I: MDS Int Rt Series Development Interest Rates since 1729 20/30CMT Avg US Long - Contemp UK Long Contemporary 16 14 12 10 8 6 4 2 0 1729 1735 1741 1747 1753 1759 1765 1771 1777 1783 1789 1795 1801 1807 1813 1819 1825 1831 1837 1843 1849 1855 1861 1867 1873 1879 1885 1891 1897 1903 1909 1915 1921 1927 1933 1939 1945 1951 1957 1963 1969 1975 1981 1987 1993 1999 2005 2011 Average 1990- 1953- 1920- 1798+ 1729+ Sources: 20/30CMT 5.39 6.21 NA NA NA FRB, H.15 Selected Interest Rates US Long 6.25 6.87 5.81 5.23 5.23 Measuringworth.com UK Long 5.49 6.64 5.58 4.66 4.42 Measuringworth.com 15

MDS Phase I: MDS Int Rt Series Development MDS Interest Rate Series since 1729 MDS Short Rate MDS Long Rate 16 14 12 10 8 6 4 2 0 1729 1735 1741 1747 1753 1759 1765 1771 1777 1783 1789 1795 1801 1807 1813 1819 1825 1831 1837 1843 1849 1855 1861 1867 1873 1879 1885 1891 1897 1903 1909 1915 1921 1927 1933 1939 1945 1951 1957 1963 1969 1975 1981 1987 1993 1999 2005 2011 Key Observations: Long Rates - floor above 2%. Current rates low, but not outside historical norms • Short rates - Only precedent for recent Short Rates is Great Depression/WWII • Short Rate volatility > Long Rate volatility • Central banks began using short rate for monetary policy after WWI. Before that, • inversions were normal; after that, rare Cycles - long and variable; long periods near lows and short periods near highs • 16

MDS Phase II: Interest Rate Data Analysis Challenges/Questions:  How to define “moderately adverse” and analyze historical data for moderate adversity?  Are New York 7 moderately adverse now? Were they ever?  Is a moderately adverse rate change the same at the short and long end of the yield curve?  Does “moderately adverse” vary with the initial interest rate environment?  What scenario shapes are suggested by appropriate analytic frameworks? 17

MDS Phase II: Interest Rate Data Analysis Solutions/Answers:  Use generally accepted CTE70 definition of moderately adverse. Considering both tails, this translates to CTE L 85 for the left tail and CTE H 85 for the right tail  Analysis of both short and long rates – no assumption of parallel shifts  Four analytic frameworks/scenario types:  Reversion scenarios  Rate change scenarios  Cyclical scenarios  AIRG scenarios 18

MDS Phase II: Interest Rate Data Analysis Traditional CTE analysis  Expected value of a variable across a tail of the distribution  Used in risk management to summarize stochastic loss results as value at risk  Under principles-based reserves, stochastic reserves use a CTE70 standard  CTE70 example: 100 scenarios with ranked losses of: $1, $2, $3,…, $99, $100. 19

MDS Phase II: Interest Rate Data Analysis MDS Empirical CTE Analysis  CTE measure applied to historical values rather than projected values  CTE measure applied to:  Interest rates  Changes in interest rates from time X to time X+t  Must consider both tails:  Because losses may result from either high or low interest rates  To achieve CTE70, BOTH tails must sum to 30%  Our analysis used an equal split: CTE L 85 (left or low rate tail); CTE H 85 (right or high rate tail) 20