Simple rules to diversify your dividend portfolio

Over the last few years, I have added layers of complexity to my diversification rules in an ongoing attempt to optimise the structure of my dividend portfolio.

I’m not complaining because I enjoy tinkering as much as the next person, but I want my investment strategy to be usable by as many people as possible, and most people only have a few hours to review their investments each week, or perhaps even each month.

With that in mind, I've taken a hatchet to my diversification process and it has now been stripped back to a handful of essential rules that always did most of the heavy lifting. Those rules are what I’m going to cover today.

But before I get into that, I should mention that this is the third in a short series of posts covering several important adjustments I’ve made to my investment process this year. Here are the first two posts:

1. How I'm hunting for UK dividend hero stocks
2. 2 Ways to value dividend-paying shares

Long-time readers may also feel a sense of deja vu as I’ve already written about diversification a couple of times over the last year or so. What this update brings to the table is a renewed focus on simplicity and it removes some of the rules added in those previous articles.

Limit your exposure to any one company

Diversification isn’t really about how many stocks you own; it’s about limiting your exposure to specific risks. So, rather than thinking in terms of how many stocks you want to own, you should think in terms of how much exposure you want to any one risk. The most obvious risk is company-specific risk, so that's where we'll start.

The number of stocks you hold is obviously related to how much company-specific risk you're exposed to, but they're not necessarily the same thing.

You could, for example, own 100 stocks, which sounds like a lot of diversification, but if 50% of your portfolio is invested in one company then half of your portfolio would still go up in smoke if that one company went bust. That isn’t what I’d call a sensible degree of diversification.

This idea of limiting your exposure to the risk of one company going bust, or even just the risk of one company cancelling its dividend, can be summed up by this simple rule:

• Rule of thumb: Keep each holding small enough so you wouldn’t be overly upset if it went bust

How much should you invest in each company?

So, take a moment to imagine that one of your investments has just gone to the wall. If it made up 1% of your portfolio, would that stop you from sleeping well at night? Would it upset you to the point where you might do something you’d later regret, like selling all of your investments and moving into cash?

With a 1% investment, most investors would say no, but what about 5% or 10%? At some percentage, you’ll start to feel uncomfortable, and that should probably be your maximum position size.

Once you’ve identified your maximum position size, you then need to think about your default position size, which has to be smaller than your maximum size. Why? Because you can’t open new positions at your maximum position size. If you did, the slightest share price rise would immediately push them over the limit, so there needs to be a gap between your opening position size (which I'll call the default position size) and your maximum position size.

This sounds simple, but it raises many questions. For example:

• How big should the gap be between the default and maximum size?
• Should the gap always be the same for every stock?
• Should the default and maximum size be fixed or should they vary over time?
• Should you overweight the highest-yielding stocks or those where you have the most conviction?

This can be a rabbit hole and over the years I’ve experimented with wide and narrow gaps as well as fixed and variable position size targets. After all of that experimentation, I now prefer simplicity over complexity and ease of use over optimisation.

Because of that, I recently went back to using a fixed default size for all holdings (also known as equal-weighting), a fixed maximum size, a fixed minimum size and a relatively large amount of wiggle room between the minimum and maximum sizes (also known as rebalancing triggers, because holdings that move beyond the maximum or minimum size trigger a rebalancing trade that trims or tops them up to the default size).

For example, I start to get uncomfortable when a position exceeds 5%. I can stomach a small degree of discomfort, so my maximum position size is currently set to 6%. On the other hand, I don’t want to waste time analysing stocks that only make up 1% of my portfolio, so I have a minimum size of 2% to avoid time wasters. This gives me a position size range of 2% to 6%.

My default size is in the middle of that range, at 4%. I recently experimented with having a 5% default size, but having a large number of holdings close to my 5% discomfort threshold made me feel like the Princess and the Pea. Like the Princess, I don’t want to be kept awake at night by something that makes me uncomfortable, so I’ve moved back to a more conservative and more relaxing default of 4%.

How many companies should your portfolio hold?

The number of holdings will be driven by your maximum and default position sizes. In my case, the 4% default size is equal to 1/25th of the portfolio, so an equally weighted portfolio with a 4% default position size will naturally gravitate towards 25 holdings. But that number doesn’t have to be set in stone. It would be entirely reasonable to maintain the number of holdings within a range, perhaps 20 to 25, 25 to 30 or even as wide a range as 20 to 30 holdings.

Having used both approaches, I do like the flexibility that comes with allowing the number of holdings to move around within a range, but I prefer the simplicity of a fixed number of holdings because it helps me to be more disciplined.

With a fixed number of holdings, you have to follow a one-in-one-out policy. If you want to sell a holding, you need to be confident that you can find a better alternative, and if you want to add a new holding, it has to be good enough so that you’re willing to sell one of your existing holdings to make room.

You can think of this as managing a portfolio using buckets. The UK Dividend Stocks Portfolio has 25 buckets and sometimes those 25 buckets will each contain a stock. At other times, 24 of those buckets will contain shares and one will contain the cash proceeds of a recent sale. If there are 25 stocks then I know my next trade will be a sale and if there are 24 stocks then I know my next trade will be a buy.

I used this approach for about ten years from 2011 to 2020 and it worked very well, but I moved away from it after the pandemic to experiment with other ways of managing the number of holdings. Having tried various other approaches, I still enjoy the simple discipline of this one-in-one-out approach, and I think its simplicity and baked-in discipline make it suitable for a lot of investors.

How often should you trade?

Although trading frequency isn't a diversification decision, it will be affected by the size of the gap between your default position size and your rebalancing triggers (the smaller the gap, the more frequently you'll have to rebalance), so it is related to the topic at hand.

Personally, I like the simple discipline of making no more than one trade each month and making that trade at the same time each month.

This reduces the amount of time it takes to manage my portfolio, but it also forces me to focus on the long term because I can only make 12 trades per year, so I'm unable to make frequent emotionally-driven trades based on the short-term news cycle. It also makes me think about each trade a little more deeply, because I have a whole month to think about whether my next trade is going to be a buy, a sell, a trim or a top-up, or perhaps it will be the other sensible option, which is to do nothing.

To summarise all of that, here are the rules of thumb I use to turn these ideas into a simple checkbox process:

• Rule of thumb: Open new positions at 4%
• Rule of thumb: If a position exceeds 6%, trim it back to 4%
• Rule of thumb: If a position falls below 2%, either (a) top it up to 4%, (b) sell or (c) wait
• Rule of thumb: Follow a policy of one-in-one-out so the portfolio always has 24 or 25 holdings
• Rule of thumb: Limit yourself to one trade (buy, sell, trim or top-up) per month

The 2% top-up rule has some nuance which is worth explaining. If a position falls below the minimum size, there are broadly three reasons and three responses:

1. Top-up: The share price of a good holding has fallen to very low levels, in which case it makes sense to invest more into it by topping it up.
2. Sell: The stock has underperformed because it's a low-quality or very low-growth company, in which case it makes sense to sell, learn the relevant lessons and reinvest the proceeds into a new higher-quality holding.
3. Wait: There’s a short-term issue that has depressed the share price and also made the stock temporarily uninvestible. This could be something like a suspended dividend or an impending legal decision which is going to either bankrupt or exonerate the business. In these cases, it may be worth waiting, even for a year or two, but only if the potential upside is attractive.

Remember, all of these rules can be adjusted to your specific situation. If, for example, you only want to invest 25% of your retirement pot into dividend stocks, you might hold 10 stocks with a default size of 2.5%, or five stocks with a default of 5%. Your maximum and minimum position sizes could then be adjusted, depending on how much trimming and topping-up you want to do.

Limit your exposure to any one sector

If you’re invested in 25 equally-weighted companies then you may think your portfolio is well diversified, but if they’re all UK clothing retailers, it probably isn’t. That’s because most clothing retailers are exposed to similar risks, whether that’s the risk of the next recession or the risk of obsolescence thanks to the rise of online shopping. This risk overlap is a feature of most companies within a given sector, so diversifying across many sectors is the next sensible step.

There are several ways to define industries and sectors. I use the Industry Classification Benchmark (ICB) system, mostly because it’s used by the London Stock Exchange and my preferred investment research tool, SharePad. The ICB system isn’t necessarily better, so if your data provider uses a different system, feel free to use it instead.

As with position sizing, I have experimented with various approaches to sector diversification, including rules to diversify by both industry and sector and rules to tilt portfolios towards defensive sectors and away from cyclical sectors. Once again, I now prefer simplicity over complexity and ease of use over optimisation, so I now use one simple rule to limit sector-specific risk:

• Rule of thumb: Don’t have more than 10% of your holdings in one sector

With 25 holdings, that limits me to two holdings from any one sector, so my portfolio will always be diversified across at least 12 sectors.

One thing I should point out is that there are five FTSE Sectors for financial stocks (Banks, Finance & Credit Services, Investment Banking & Brokerage Services, Life Insurance, Non-life Insurance), so if you had 10% in each of those sectors you would have 50% of your portfolio in financials. That could be a problem because UK financial firms are all exposed to the same regulatory risk from the same regulator (the FCA), which makes diversification across those sectors less effective at reducing risk.

Over the last year or so, I’ve used a rule to limit my portfolio’s exposure to the financial industry (or any one industry) to 20% at most, which is sensible. However, that rule is restrictive, it adds another layer of complexity and, given the low probability of finding multiple high-quality dividend hero stocks in every financial sector, it may be trying to avoid a risk that is, in reality, vanishingly small.

And so, in the name of simplicity and flexibility, I no longer use that industry diversification rule and instead focus exclusively on the easy-to-implement 10% per sector rule.

Limit your exposure to any one country

Spreading your portfolio across many different companies and sectors is a good start, but if most of those companies generate most of their revenues in the UK, the portfolio could be overly susceptible to the ups and downs of that country's economy.

Fortunately, it’s easy to get around this problem by investing in companies that generate much of their revenue and profit from outside the UK.

In most cases, you can find a company’s geographic exposure in its latest annual report, or you can use figures from data providers such as Morningstar or SharePad. If you can’t find a figure for UK revenues then you could use a reasonable alternative. For example, a company might publish revenue figures for Western Europe, and you could then make a ballpark guess as to its UK revenues based on that.

Once you know (or have estimated) the UK exposure for all of your holdings, you can apply a simple rule of thumb to limit your geographical risk. Here’s my rule:

• Rule of thumb: On average, holdings should generate less than 50% of their revenue in the UK

There are various ways to calculate this and in my case, I just calculate a simple average across all of my portfolio’s holdings. When I’m looking at my next potential investment, if its UK exposure would be enough to push the portfolio’s exposure above 50%, I’ll move on and look for a more internationally focused alternative.

As with all of the previous rules, the precise geographical limit can be adjusted based on your preferences.

For most people and most portfolios, simple beats complex 

If you enjoy process-driven investing as much as I do then it’s easy to keep layering new rules and procedures on top of one another. There isn’t necessarily anything wrong with that, but there are benefits to stripping away the superfluous until there’s nothing left but the essentials.

Although the simple diversification rules above may not quite be down to the essentials, I think they’re close and they’re certainly a good place for most dividend investors to start.