A financial enigma
The investment and academic worlds are united, generally speaking, on what dividends are, where they come from, and even how large they are going to be next year. They are paid to shareholders for their equity investments and “they are roughly 5% per annum” – as one recent top practitioner said, looking slightly bemused, to some academics trying to discover something, anything, related to this rather strange financial concept, after all. There is a tide of academic literature devoted to the question of what signal is conveyed when dividends are announced. Yet there is very little to tell us what dividends actually are and what characteristics they have over time. Occasionally glimpses of hope are given to us by the great lighthouse keepers of finance. So one thing we do know, from Robert Shiller the Nobel laureate for Economics in 2013, is that the volatility of dividends is at odds with the volatility of their corresponding equity shares. In fact, the volatility of dividends seems to be over five times lower. Why is dividend volatility so low?
This puzzle has not been resolved in the academic literature and it is far too ambitious to solve here. But it is not only of academic interest, given EUREX’s newly opened market in dividend derivatives. Derivatives feed on volatility and on the surface it may seem paradoxical to have a thriving market based on an asset exhibiting historically low volatilities. So what lies beneath?
The reason behind the low volatility is very simple, in my opinion, and it is largely misunderstood both in academia and in practice. While the share of one company represents a stock-type of variable (witness the convention of describing shares of equity as stock), their dividends are actually flow-type variables. The value of the shares of global commodities trader Cargill, for example, is ultimately determined by the views of equity market participants. Cargill’s price goes up when more investors want to buy the share, and its price goes down when more investors want to sell the share. At the same time, its dividend – calculated in dollars per share – is decided by the executive team of the company and not by equity market participants. For Cargill’s dividend, a jump up or a jump down, under the same market conditions, may occur if there is a change in the executive team. Some might argue, then, that there should be more volatility associated with the value of dividends, particularly when considering a long investment horizon, say of 30 years.
The share price of the company may go down and yet the dividend amount may go up if executives want to use this channel to calm down the markets about the value of the company. However, the same dollar amount of dividends may represent very different percentages of the share price, since the payment will be done in the future. A dividend of $10 represents 5% if the share price at the time when the dividend is paid is $200 but it represents 10% if the share price is $100. This, then, is a flow-type variable. Furthermore, dividends are paid only a few times, mostly once per annum and towards the end of the financial year. For the majority of companies there is only one reported value of its share price cum-dividend within the entire year. With very few values reported, the volatility of cash dividends can be only low. Plus, as a proportion of the value of the company, they represent a minute fraction and this imposes an upper boundary that will inhibit the dividends growth over time.
Reaping the dividend
To my mind, dividends are a by-product of shares but they are formed beneath the surface hidden away from the public. They can be also understood as an individual asset class, similar to coupons on coupon-bearing bonds, or rents from commercial properties.
Dividends are embedded nowadays in complex financial contracts such as structured products. These contracts are for savvy investors and the entire play may be actually on tax since the same contract may attract a high level of tax in one country (UK) and a low level of tax in another (Holland, Switzerland). If death and tax are the only two certainties in life, it seems that we are equal only in respect of the former!
Dividends are also important for pricing equity options. In this context, dividends cause serious problems to financial engineers. In spite of slow but periodic production, dividends are stochastic. We do not know exactly when they are going to be paid or how large they are going to be. Consequently, the models for pricing equity options that assume a constant dividend yield will produce unreliable results.This is, coincidentally, an area of business where banks do not let auditors or regulators look.
With the introduction of dividend futures on EUREX (based on ten year maturities for the pan-European STOXX50® equity index) some standardisation of the information about dividends is taking place. The futures contracts allow the investor to gauge the forward-looking market view on the total sum of dividends that the top companies may pay each year. Unfortunately, the modelling of dividends for pricing dividend derivatives purposes is still in its infancy and more research is needed to help bankers develop this market further. Some recent research at the Centre for Quantitative Finance Research may be able to contribute to the conversation. There are also European options (call and put) traded on these futures so all types of investor can join the play based on their risk profiles.
It is not surprising that hedge funds and private equity investors are considering dividends via derivatives at a time when nominal interest rates are going into negative territory. There are no negative dividends (yet?!) so during these extraordinary times, dividend yields of 3%-6% look very attractive.
Going with the flow
Many questions for the future are still unanswered. With the request by regulators to increase dramatically their equity capital, will banks be able to pay dividends similar to what they paid in the past? If they pay less is that a signal that their financing is tight and their future performance will deteriorate? In these circumstances they may pay more dividends than before in order to fence off these worries by investors. Is there a cross-country dividend trade between US and UK or between US and Europe? What happens if the interest rates become more negative and stay there for some time? Is this an opportunity to go long or short futures on dividends?
There is some concern that trades on dividends of individual companies are intrinsically illiquid and these positions will be very difficult to hedge. While this is true there is also a solution available – cross-hedging. The dividend index constructed from the dividends of those companies comprising the STOXX50® index (and using the same weighting scheme) has liquid dividend futures and can be used for hedging individual companies dividends.
The beauty of being able to trade futures with ten year maturities is that an investor can trade along the curve. You may say this is for the rocket scientists but I disagree. It is conceptually no different than trading the shape of the forward curve for oil price or real estate. And the prices are those offered on the exchange so actually you may not need to do any heavy maths, just have a good view of where the economy is going near term and mid-term. If you think that the technological revolution is only beginning then you may like long dividends between 3 and 5 years, so you need to go long 5-year futures and short 3-year futures on the same dividend index. In plain English, you pay nothing now, you do nothing or nearly nothing for the next three years, and then wake up to see whether you are going in the money or not. I do not know about you but I think we should go with the flow.
Professor Radu Tunaru is currently Head of Finance group in Kent Business School, and Director of the Centre for Quantitative Finance Research (CEQUFIN).