Playbooks / Portfolio
What diversification does, and does not
The rule
Diversification is the one thing in investing that is genuinely free, and it is routinely spent on the wrong axis.
The two risks it separates
Specific risk belongs to one holding. A tenant defaults, a chiller fails, a company loses a lawsuit. Spreading across enough holdings makes any single one immaterial, and this is the risk diversification eliminates.
Systematic risk is shared. A recession, a rate shock, a regional downturn. It affects everything at once and no amount of spreading removes it, because there is nothing to spread into that is not also affected.
The number of lines in a portfolio addresses the first and says nothing about the second.
Where the count misleads
Thirty holdings in one country's equity market removes specific risk almost entirely and leaves you fully exposed to that country. Six apartments in one city is one property market with six sets of paperwork. Five funds that all hold the same large global companies is one fund with five fee schedules.
The useful test is not how many things do I own. It is how many different conditions do my holdings need in order to work. That is the four boxes question and it counts economic environments rather than line items.
The diminishing returns
Most of the specific risk reduction from adding holdings happens early. Going from one holding to ten removes the great majority of it. Going from thirty to a hundred changes very little.
Which means the marginal effort is better spent on the second axis: different economies, different currencies, different asset classes, different drivers. A portfolio of twenty holdings across four genuinely different exposures is better diversified than a hundred holdings inside one.
What it does not do
It does not prevent losses. In a systematic decline a diversified portfolio falls. It falls less than a concentrated one, usually, and that is the whole claim.
It does not improve expected return. It improves return for a given level of risk. Those are different statements and the second is the honest one.
It does not survive correlations changing. Assets that behave independently in calm markets frequently move together in a crisis, which is when the diversification was supposed to earn its keep.
The version that applies to you
For most expatriate investors the binding constraint is not the number of funds. It is that salary, property, currency and often residency are all pointed at one place. That single fact dominates everything happening inside the investment account.
Fixing it means adding exposures that do not depend on the same conditions, which usually means assets outside the region and outside the currency, and it is the reason concentration limits and currency risk sit next to this page rather than apart from it.
The arithmetic
Where it breaks
- It cannot remove systematic risk, and portfolios sold as diversified frequently imply protection they cannot deliver in a broad decline.
- Correlations rise in crises, so the measured diversification of a calm period overstates what is available in a bad one.
- Over-diversification has real costs in complexity, fees and the temptation to tinker, without adding meaningful risk reduction beyond a point.
- It says nothing about valuation. A well diversified portfolio of expensive assets is still a portfolio of expensive assets.
- Diversifying into things you do not understand in order to raise the count is worse than a simpler portfolio you can hold through a bad year.
- For most people the largest single exposure is human capital, the ability to earn, and no investment portfolio diversifies that.
When to use it
When a portfolio feels diversified because it has many lines. Count conditions rather than holdings, and the answer is frequently uncomfortable.
The Safe Withdrawal Rate does this arithmetic for you, in your currency, in about thirty seconds.
Open the calculatorSources
Last reviewed . Educational research, not personal advice. Disclosure standards.