DISCK/DISCA Share Class Arb
DISCA: $36.29
DISCK: $31.48
Current Spread: 15.3%
Thesis:
Long DISCK, short DISCA on a dollar-weighted basis. DISCK currently trades at a spread of 15% vs. DISCA. This compares to the historical spread of 0-10% depending on your time frame and the current CMCSA/CMCSK spread of 5%. The spread has widened recently due to the Russell rebalancing. The catalyst to reverse this spread back into ~10% or tighter now that the rebalancing is done is the high likelihood of Discovery initiating a share repurchase plan on the DISCK shares. We expect to see an announcement of a share repurchase program in the next 3 months and we think there is a ~4.5% gain in the spread tightening in the next 3 months or a ~19%+ IRR with very little risk.
Description:
Discovery Communications creates content for several cable channels including Discovery Channel, TLC, and Animal Planet. Discovery has a voting share class (DISCA) and a non-voting share class (DISCK). John Malone and partners controls the vote so there is little value of a voting share vs. a non-voting share. DISCA is 3x more liquid than DISCK but DISCK still trades ~$25 mm per day.
Valuation & Analysis:
The link below to a word doc shows a 2-year chart of the spread between DISCK/DISCA. You can see that the share classes traded between a 0-5% spread until Q1 09, widened out and wrapped around 10% from Q2 09 to Q3 09, and have slowly widened from 10% to a high of 21% over the last 9 months, with a recent acceleration in the past 3 months from 15% to 21% in anticipation of the rebalancing effect. The spread has tightened back into 16% with the cessation of rebalancing flows but we think there is room for continued tightening back to a historical/average of less than 10% especially on the back of a DISCK share repurchase.
http://www.mediafire.com/?nymqmdlwnlz
That same word doc has a scenario analysis. In the base case we assume that the spread reverts back to its historical average of ~10%. This results in a net gain of $900k or 4.5% on a $20 mm position. The low case imagines a correlation=1 market stress event that blows all arbitrage positions wider; assume the spread widens to 20% which results in a loss of $850k mm or 4.3% on a $20 mm position. The high case assumes the 2 share classes tighten to parity which is what we saw in 2008; however we would be out of the trade well before then.
We think there is a DISCK share repurchase coming because 1) Discovery is under-levered at 1.75x net debt to EBITDA vs. management guidance of 2.5-3.0x and has excess cash flow being generated by the business 2) There are reports of John Malone discussing the discount and that it was a no-brainer for Discovery to be repurchasing shares of DISCK 3) Sell side analysts are beginning to put out notes (ex. MS note from July 8th) highlighting the discount between the K shares and the A shares and talking up the potential for a DISCK share repurchase.
Risks:
DISCA and DISCK represent the same economic interest with a different voting structure. Thus the risk to the trade is not one of fundamental change in the business but rather market volatility. We will lose money if we unwind the trade through a market stress event that widens the spread of all arbitrage positions. Assuming we blow out at a 20% spread we would lose $860k or 4.3% on a $20 mm position. If we get clarity that there is no DISCK share repurchase coming then the catalyst for spread tightening would be removed and we should exit the trade.
Catalysts:
Russell rebalancing effects are done and DISCK share repurchase by the company.
Timeline:
Expect an announcement of a DISCK buyback in the next 3 months. Expected holding period of 0-3 months with a target spread of the historical average of ~10% to begin unwinding the trade.