The core assumption of Matching Adjustment is that default risk changes very slowly over time, which is why the average spread over a 30 years period is one of the inputs into the Fundamental Spread (=credit risk spread) calculation. It follows that sharp movements in bond spreads cannot be the result of changes in default risk, hence cannot correspond to any change in the credit risk spread. The change must therefore be in the Matching Adjustment (=liquidity) spread.
How could we test this assumption?
One way would be to construct an illiquid but risk free bond, which would be a perfect, though hypothetical, hedge for the illiquid annuity portfolio. We could do this by purchasing an illiquid but risky bond, let’s say on Ford Motor Co, then purchase credit default insurance in the CDS market, which will hedge our credit risk only.
See the chart above for Ford bond spread against the CDS spread.
But, um, this isn’t quite the result we wanted. The CDS spread seems to follow the bond spread all the way up during the crisis, and there is even a short period where the CDS spread is greater than the bond spread! Hence negative illiquidity! If we hedged the bond exposure with the CDS, we would find our perfectly riskless but illiquid bond has a seemingly considerable risk.
How could that be? Our contact form is here.