Thinking Beyond The Traditional Risk Management Programs Sheldon - - PDF document

Equity-Based Insurance Guarantees Conference

• Nov. 5-6, 2018

Chicago, IL

Thinking Beyond The Traditional Risk Management Programs Sheldon Epstein

SOA Antitrust Compliance Guidelines SOA Presentation Disclaimer

Sponsored by

Proprietary Reinsurance Solutions

Thinking Beyond The Thinking Beyond The Traditional Risk Traditional Risk Management Management Programs Programs Sheldon Epstein Sheldon Epstein

A CONFIDENTIAL PRESENTATION EBIG 2018

• 5 November 2018

Session 1A (1045 – 1215 hours)

2

Aga m Ca pita l Ma na gement

This confidential preliminary information has been prepared by Agam Capital Management (“Agam”) solely for information purposes and is being furnished solely to assist the recipient in deciding whether to proceed with further analysis of the transaction contemplated herein. This document does not constitute an offer or invitation for the sale or purchase of securities. The information set

• ut herein is preliminary and should not be relied upon for any purpose. does not make any representation or warranty, express or implied, as to the accuracy or completeness of the information

contained herein and shall not have any liability for such information. Interested parties should conduct their own investigation and analysis of Agam, its business, prospects and prospective results of

• perations and financial condition. Except for internal use, this information may not be excerpted from, summarized, distributed, reproduced or used without the prior written consent of Agam. The

investment opportunity described herein is speculative and entails a high degree of risk. Due to the illiquidity of this investment, if you invest, you must expect to bear the economic risk of the investment for an indefinite period. Agam expects that no market will develop for the potential investment opportunity described herein. Certain statements in this document that are not historical fact constitute "forward-looking statements." You are cautioned not to place undue reliance on these forward-looking statements. Agam generally identifies forward-looking statements by using words like "believe," "intend," "target," "expect," "estimate," "may," "should," "plan," "project," "contemplate," "anticipate," "predict" or similar expressions. You can also identify forward-looking statements by discussions of strategies, plans or intentions. Such forward-looking statements involve known and unknown risks, uncertainties and other factors, which may cause the actual results of Agam to be materially different from historical results or from any results expressed or implied by such forward-looking statements. All forward- looking statements herein are qualified in their entirety by this cautionary statement. Agam made the statements in these materials as of the date hereof unless it is stated otherwise. Neither the delivery of these materials, nor any sale of securities by Agam after the date of these materials, shall create any implication that the information contained herein or the affairs of Agam have not changed since the date hereof or that such information is correct as of any time subsequent to its date. Agam management based all estimates and projections as to events that may occur in the future upon their best judgment as of the date of these materials and upon assumptions and circumstances and events that have not yet taken place, may not have an empirical basis, are subject to variation and are inherently unpredictable. Whether or not such estimates or projections may be achieved will depend upon Agam achieving its overall business objectives and the availability of funds, including funds from the sale of securities. There can be no assurance that any estimates or assumptions will prove accurate or that any of the projections will be realized. Agam does not guarantee that any of these projections will be attained. Actual results will vary from the projections, and such variations may be material. You should not construe the contents of these materials as legal, tax or investment advice. You should consult your own counsel, accountant or business advisor as to legal and other matters concerning any investment decisions. These materials are not all-inclusive. Nor do they contain all the information which you may require. Consult your own legal, tax or investment counsel regarding the legality or suitability of your investment under applicable legal, investment or similar laws, regulations or fiduciary standards. Agam does not make any representation regarding your investment herein under any legal, investment or similar law, regulations or fiduciary standards. The information in this document is not targeted at the residents of any particular country and is not intended for distribution to, or use by, any person in any jurisdiction or country where such distribution or use would be contrary to local law or regulation. Furthermore, no investment opportunity will be extended to persons resident in any jurisdiction or country where such distribution would be contrary to local law or regulation.

Importa nt Disclosures

3

Introduction

• Overview of risk-management strategies deployed by Variable Annuity (VA) companies

to hedge their riders

• Current alternatives to the predominant strategies
• Innovative and practical ways of managing VA contract risks more effectively in a cost-

efficient manner.

4

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• Overview
• In Dec. 2016, Milliman published “How effective is variable annuity guarantee hedging?”
• They defined “Hedge Effectiveness”
• Effectiveness of loss recovery can be measured by the ratio of the change in hedge asset

value to change in liability increase during a market downturn.

• Definition is focused on GAAP P&L volatility rather than targeting other metrics such as

statutory solvency

• In particular the general observation is that most companies target hedging VA guarantees

that fall under FAS 133/157 since they respond to capital market inputs,

• The most common approach is to employ a “2-Greek” strategy (Delta and Rho)
• When listed equity options are utilized, companies tend to position shorter-dated options

than would be derived from liability sensitivities

• FASB long-dated contract improvements would move all of the VA guarantees to FAS 133/157

Dyna mic Replica tion

5

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• Risks
• Basis Risk
• Gap Risk
• Vega Risk
• Estimation of higher order Greeks
• Cross-Market Correlation
• Statutory vs GAAP vs “Economic” mismatches

Dyna mic Replica tion

6

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• Objective: Minimize the volatility of the underlying separate account values by enforcing

stylistic or algorithmic allocations

• CPPI
• Investment allocation constraints
• Managed Volatility Funds
• These approaches rely to some extent on
• positive correlation of rates and equity price movements;
• algorithmic approaches also rely on the observation that implied/realized volatility increases with a large

directional move

• Two observations
• Volatility Managed Funds do achieve underlying target volatility
• Algorithmic approaches tend to ignore observed mean reversion in price movements serially over time

leads to a whipsaw effect potentially reducing future fund growth

Investment Constra ints a nd Ma na ged Vola tility Funds

7

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• A number of public VA writers have disclosed in their earnings disclosures and SEC

filings that they are targeting statutory capital either primarily or secondarily

• 1. “In addition to our dynamic hedging strategy, we have recently implemented static hedge positions

designed to mitigate the adverse impact of changing market conditions on our statutory capital. ”

• Dynamic Replication with Statutory Overlay
• 2. “We focus our hedging activities primarily on mitigating the risk from larger movements in capital

markets, which may deplete variable annuity contract holder account values, and may increase long- term variable annuity guarantee claims. “

• Targeting Long Term Solvency and not short term volatility
• 3. “Certain of our products, particularly our variable annuity products,…,we may choose to hedge these

risks on a basis that does not correspond to their anticipated or actual impact upon our results of

• perations or financial position under U.S. GAAP”
• Keep an eye on GAAP but target enterprise risk
• 4. AGAM: Target statutory solvency within variable annuities and economic value as an overlay across all

products

Ta rgeting non- GAAP metrics

8

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• There historically have been providers of reinsurance for VA guarantees
• Pre-2009 a major reinsurer existed for GMDB and GMIB riders
• From 2012-2016 reinsurers provided coverage for GLWB and GMDB riders
• Most of the providers (and captives which took the same approach) seem to have been

located offshore where the local statutory basis would have allowed them to use U.S. GAAP for valuation purposes

• SOP-03-1 was effectively the statutory valuation basis
• Pre-2009 reinsurers stopped providing coverage as it was apparent (at least to those

writers) that they were not valuing and/or hedging all the market risks

• The more recent generation of reinsurers have stopped providing reinsurance on the

existing US GAAP basis

• FASB Targeted Improvements for Long Dated Contracts
• NAIC reforms both for captives and for VA

Reinsura nce of VA Gua ra ntees

9

Overview of risk- ma na gement stra tegies deployed by VA hedgers

• Transactions in the last 24 months where VA writers sold/spun-off entities that had

substantial VA exposure

• Facilitated by emergence of new alternative capital providers (PE)
• Part of strategic refocus of companies to fee-based, lower capital market risk, less

capital intensive product

• “Talcott Resolution”/Hartford
• Voya
• AXA spin off of its U.S. retail operations
• MetLife spin off of U.S. retail life and annuity business (including VA)
• Implicit in either the equity price movement or the transaction prices was a cost of

transferring the VA at CTE95-CTE98 level

• Still in almost every case the equity prices of these companies increased in the

immediate aftermath of the transactions

Ca pita l Ma rket Sa les or Spin- Offs

10

Ta rgeting Sta tutory Solvency a nd Entity Va lue

AG-43 has 3 main uses of the term “effectiveness”

• 1. When modeling CDHS, how effectively do you adhere to the strategy
• 2. If taking credit for a CDHS without modeling it, effectiveness refers to how well did the hedges work out
• 3. Calculating the overall requirement: The reg defines ‘an “effectiveness factor”’ E is used to weight the

best efforts requirement (incorporating the CDHS either through 1 or 2 above) vs the requirement of just running off existing hedges

• “Reflects actuary’s view as to level of sophistication of the model and its ability to properly reflect the

parameters of the hedging strategy” CDHS in VM-21/AG-43/C3P2

• The model will determine what reserve/capital is needed as a result of the hedging strategy
• A good strategy, modeled well, reduces the amount of capital needed
• A bad strategy, modelled well, will not reduce the amount of capital needed
• A good strategy not modeled well, will not reduce the amount of capital needed
• You need to prove you follow the strategy in real life
• The choice of “Hedge Target”/strategy is left up to the company

CDHS a nd Hedge Effectiveness

11

Ta rgeting Sta tutory Solvency a nd Entity Va lue

• ENTIRE contract is considered in CDHS as in Statutory calculations
• M&E, Rider Fees, NRSI
• Trails, Admin, etc.
• All guarantees, DB, AB, IB, WB
• C3P2/RBC: TAR in excess of the statutory reserve
• Equally effective Current and Future AG43/C3P2
• If GPVAD were equal in all scenarios then CTEx=CTEy for all x,y and CTE0=CTE70=CTE99
• Reduce computation by developing a real hedge target that is a proxy for CTE0
• Calculate hedge target Greeks (delta/rho) inclusive of:
• (starting) assets,
• reinvestment strategy
• Rho hedging is minimized if asset reinvestment strategy is considered first level rho-hedging
• NAIC reform provides a safe-harbor for delta/rho hedging
• Must actually follow the CDHS, executing delta/rho in real life
• Modeled “Hedge Target” exactly as we define it

Modeled CDHS Redefined

12

Sta tutory CDHS Effica cy

Lega cy GMIB – Impa ct of Sta tutory Ba sed Hedging Progra m on Tota l Assets Requirement

100.00 207.00 118.72 179.25 113.28 163.34 139.27 127.50 25.00 50.00 75.00 100.00 125.00 150.00 175.00 200.00 225.00 30 40 50 60 70 80 90 100 CTE Value (CTE70 no CDHS =100) CTE Level

Impact of simple CDHS Hedging Target Proxy on TAR

(1) No CDHS (2) Simple CDHS Targeting CTE0 (3) Better CDHS Targeting CTE0 Negative Cede with (1) Negative Cede with (3)

CTE70 CTE98

• CDHS (3) allows for a reduction of negative Cede by ~12% relative to original CTE70 with no CDHS
• This is a result of the impact of reducing CTE98 by ~44% relative to original CTE70 with no CDHS
• Ceding company frees up capital of ~80% of CTE70 by reducing CTE98 TAR of 207 in return for negative cede of 127.5

Negative Ceding Commission

13

Entity Va lue Hedging

How does it fit in?

VM-21 explicitly does not allow for inclusion of cash-flows from products not included under the reg:

• “The use of products not falling under the scope of these requirements (e.g., equity-indexed annuities) as a hedge shall

not be recognized in the determination of accumulated deficiencies.” Implication:

• Recognition that there may be portfolio and other effects, but they don’t increase/reduce the requirements
• As long as the company holds adequate aggregate reserves and capital derived it may manage the entity to achieve other

targets To minimize aggregate market based hedging costs, it makes sense to aggregate and execute entity level “hedges” net of product level hedges

• A super-solvency criterion is to
• minimize aggregate statutory solvency volatility subject to constraints on economic (or other management level

targets)

Entity level management encompasses all exposures net of product level hedges – longer term less liquid hedges (options etc.) at this level e.g. sum(MRN)-sum(Product level) exposure

14

“Modified” Risk Neutra l vs Sta tutory Va lua tion

Compa ring ma rket models

“Modified” Risk Neutral

• RN seeks to match market prices of securities – arbitrage

free

• Shortcuts abound, all of the price is assumed to be

accounted for by the market model

• More refined includes the cost of capital, correlation and

non-continuous markets

• Underlying AV (after survival) ~

(1 + RF+v(market)*Y*sqrt(t)- fees)^t / (1+RF+ v(rf)*X*sqrt(t) + CR-prem)^t On 8/31/18: RF=~3.25%, v(equity) ~20%, v(rf)=~100bp and correlation Y|X=~25%

• all are dynamic

Statutory

• Calibrated to long-term historical behaviors rather

then replication prices

• Implicitly embeds cost-of-capital the determination of

GPVAD

• AAA generator has zero rate/equity correlation and

100% continuous markets

• Underlying AV (after survival) ~

(1 + 7.5% + 15%*Y*sqrt(t)- fees)^t / (1+MRP+ v(MRP)*X(implied) + CR-prem)^t On 8/31/18: MRP=~3.50%, v(equity) ~15%, v(MRP)=~100bp and correlation Y|X=0%

• all are sticky

15

“Modified” Risk Neutra l vs Sta tutory Va lua tion

Compa ring ma rket models “Modified” Risk Neutral

DELTA = $6.7 S&P 500 RHO = $42.4 20 year UST

• MRN has more rate exposure, less equity

than STAT

• MRN entails utilizing more interest rate
• ptions and fewer equity options than

STAT = more Flexibility

Statutory DELTA = $8.1 S&P 500 RHO = $13.0 20 year UST

• Co-calibration and correlation of rates with equities allocates some of the equity

volatility/impact to rates, reducing equity impact and increasing rate impact in MRN vs STAT

• MRP as a slow moving average of historical rates dampens impact of rates on STAT further

relative to MRN

• Legacy GMIB book, AV=$100, rho in excess of reserve invested in $20 20yr UST equivalent

16

La test a pproa ch to VA Risk Ma na gement

• Quota-share Reinsurance of the entire contract (legacy or new business)
• Reinsurer can price for maximum efficiency across all of the product components vs rider only
• Allows for offsetting actuarial assumptions as well as capital market risks
• Free up capital required even after management of risk to deploy in other strategic
• pportunities
• Equity value creation via reduced capital market uncertainty premium for cedant
• Reinsurer may be better positioned to manage risk due to various factors
• Public vs Private Equity
• Diversification across a wider range of insureds and product variations
• Economy of scale of infrastructure to manage complex legacy risk
• 2018 market environment (close to all time highs in equities and relatively interest rates post-

crisis) has reduced the differential between the cost to defease the capital market risk and statutory requirements

• Partner with reinsurer to optimize inforce management

VA Reinsura nce 2.0

17

Annuity Reinsurers

• Large, highly rated diversified reinsurer who would entertain VA part of an unhedged

portfolio of risk

• Asset aggregators who will write annuity reinsurance (predominantly FA/FIA without

GLWB) to increase AUM with some hedging

• Reinsurers who focus on risk management that leverage analytical skills and technology

to achieve their ROE targets

Differentia tion