Hey all,Do you have any ideas on how dividend swaps are traded? What factors would drive the dividend assumption embedded inside those swaps? Would appreciate if you guys could shed some lights on it. Henry Potter

You promise to give me your dividends. I pay you some discounted amount (with risk baked in). You take the money and buy a call option (from me, of course) on the stock. I make (just for you, my friend) a special price on that option. This is essentially a tax issue. Certain jurisdictions regulate this sort of thing.

Hi PaperCut,Thanks for the reply. It looks like it's using dividend amount to swap call option? People buying those dividend swaps are just for tax purposes? Do you know where I could get Dividend Swaps quote for different years ? I could not find them on Bloomberg. Thanks,Henry Potter

Almost all div swaps I've seen are OTC, and you wouldn't find them on BBG. Easiest way to "price" them is to back out the dividend yield implied in the options (synthetic forwards) and futures prices.Taxes are a main motivation, but on indexes, they can also be a simpler way to hedge a large block of dividend risk for a relatively reasonable price, but I don't see them that often. On single stocks there is a potential moral hazard if I call you to receive dividends on XYZ, then call my buddy on the board of XYZ and tell him to vote to triple the dividend payout of XYZ over the next year (or pay a "special" dividend, depending on the swap terms).

Hey Exotiq,You mentioned about backing out the dividend yield from the option prices. That means I have to know the implied volatility assumption as well. Dividend yield and implied volatility have to go hand in hand in order to make the price consistent. In a Black Scholes framework, it seems that it will be better to have a dividend term structure model to take care of options with different time to maturity. Do you have any idea how to construct a dividend term structure model ?Thanks for your input,Henry

Make a forward rates curve of dividend yields the same way you make a forward Libor curve.

QuoteOriginally posted by: henrypotterYou mentioned about backing out the dividend yield from the option prices. That means I have to know the implied volatility assumption as well. Dividend yield and implied volatility have to go hand in hand in order to make the price consistent. No, you don't need to know the implied vol: just use a vega-neutral option position like a synthetic forward (long call, short put, same strike), which gives you the effective forward price of the underlying. In fact, CBOE defines the forward price at the strike at which the price of a put and call are the same.

Hey Exotiq,I see. Anyway I have to know the options price for both put and call at the same strike. I could use put call parity to back solve the dividend yield. It is going to work for some short-dated options. However, for long dated options, it seems that you can't find the price of the options and thus you can't back solve to get the dividend yields. Am I right ? That means the dividend yield term structure could only go from time 0 to at most 3 years?Henry Potter

Hi Paper cut,LIBOR curve could be constructed using swap rates which range from 2 Year to 30 Year. However, for dividend yield, how could I find such a long dated option to back solve the dividend ? Isn't long dated options are unobservable (OTC product)? Henry Potter

Henry-I've seen dividend swaps on major indexes quoted out as far as 4 1/2 years, but not beyond that. OTC options beyond that term tend to have a bid-offer large enough to make the arbitrage-free range of div yields not quite as meaningful. For single stocks, you might expect at the very most about half that tenor, but I'm sure there are exceptions. Internally, I've seen banks extrapolate dividend curves out to 30 years, because they need to price profitable dividend trades. It's a similar problem to quoting a 100-year interest rate swap...

Hi Exotiq,I am more interested in knowing how to construct the dividend term structure on index (SPX/NDX). Internally, how do banks project the dividend beyond 4.5 years ? I have no idea how they project interest rate to 100 years. I actually heard that they just assume dividend yield be constant in long term. I may be wrong though. Well, even they don't have to price dividend trade, they should still use dividend assumption to calculate the options prices on their books, right? In short, it seems that I have to get put, call prices for options from 0 to 3 years and project it forward with whatever assumption I would like to make ? Sorry for my ignorance and I appreciate your help. Henry Potter

I expect you'll find it difficult to get decent prices for individual stock options beyond 3 years, so backing out the longer term divs from market data will be next to impossible.You'll have to fall back on estimating the divs beyond the point for which you get quotes, using some guesstimate of growth rates. You will, of course, be able to get index option prices for much longer than 2-3 years so you can, for a basket of shares, replicate the index and see if your replicated forward on your basket agrees to your index forward and use this to calibrate your growth assumptions.

QuoteYou will, of course, be able to get index option prices for much longer than 2-3 years so you can, for a basket of shares, replicate the index and see if your replicated forward on your basket agrees to your index forward and use this to calibrate your growth assumptions. How could I get index option prices much longer than 2-3 years on exchange ? I don't quite follow your growh assumption calibration method. Are you suggesting me to implement a growth assumptions on the current dividend of the index to best fit the dividend yield that backs out from 1-3 years index option ? How do banks model the forward dividend yield assumption? I know there is another thread dicussing the discrete dividend problem on BS pricing but my question is more related to the projection of the dividend. SPX Dividend yield assumptions have been up lately. Anybody has any idea why this happened ? Many Thanks!

you can get tradable prices on SPX combos up to 10yrs by calling any decent broker. a combo is call-put at same strike. they are quoted with a reference swap rate so backing out a divyield is trivial.other indeces will be harder to go out that far.how to extrapolate divyields to maturities longer than available markets? that is where you as a trader have to take a view, like everybody else.

QuoteOriginally posted by: granchio a combo is call-put at same strike. they are quoted with a reference swap rate so backing out a divyield is trivial..Why is backing out dividend yield trivial ? If I use their implied vol surface, the only unknown is dividend yield, isn't it ? I thought they are just quoted by % of the notional. Looks like I could only use broker's marks at a certain point and try to back out the dividend yield. Then do a fitting on those dividend yield.