Chart 27: SX5E 2Y var convexity (var strike vs. ATMf vol) has been driven lower by (A) a decline in the Tail liquidity Risk Premium and (B) volatility skew becoming more linear (i.e., less convex in strike) (A) VarSwap - 1dStrip TRP ayy ——(B) VarSwap - Linear skew approximation* We recently noted that ESTX50 long-dated (2Y) variance | (A) + (8) convexity (var strike vs. ATMf vol) had declined considerably in ———VarSwap - ATMf vol | 1 ‘ F er P WW recent months. This has been driven by (A) a decline in the Tail 6% j N Vv liquidity Risk Premium (as defined in our piece More to variance fs “\ swaps than meets the eye) and (B) volatility skew becoming 4% L he y ‘A more linear (i.e., 80-90 skew being similar to 90-100 skew). . at seal: pte | : Both these components of variance convexity suggest that \ ower’ wy T long-dated ESTX50 tails {e.g., VarSwaps or far OTM puts) are 2% ath, ath historically cheap vs. ATMF vol. 0% AAA N oD oD oD t+ — ike) ike) Ww Oo © oO SS MN ee >+ee tees aeaei acs o oO oO o oO oO oO oO oO o oO oO oO oO oO na or roonor et wnonor et norx FF )0 979 FE Source: BofA Merrill Lynch Global Research. Data: 21-Sep-16 to 31-May-17. *Linear approximation of variance strike = ATMf vol squared x (1 +3 x T x skew squared), where Tis time to maturity and skew = [implied vol at strike 90 - implied vol at strike 100 ]/[ 90 - 100]. For more details see Derman’s 1999 paper: More than you ever wanted to know about volatility swaps. Chart 28: GBPUSD short-dated (1 wk) implied has reached its 92 4yr percentile ahead of the UK snap election on 8-Jun. It is, however, still at - f Jun-16 Jan-17 Brexit ; ; ; Seems Sasvirate Ng uptere court With the UK parliamentary elections less than one week away, 35 May-15 UK rulling short-dated (1wk) implied vols on both the FTSE and GBPUSD 30 general election Weare appear more elevated vs. history, compared to last week. GBPUSD 1wk ATM vol is exhibiting some concern as it is a | trading in its 92"¢ 4yr percentile. It is, however, still