Transcript
Most investors can name a handful of financial products. Far fewer can explain what those products actually are — meaning, what economic relationship they represent and how that relationship earns a return. That gap causes real problems. People buy an ETF thinking they have diversified. They put money into superannuation and assume the wrapper is doing something protective. They hear private equity or crypto described as "different" from shares and bonds, without anyone explaining what different actually means. This episode maps the mechanisms that drive investment returns — and only those mechanisms — because that map applies to almost any financial product you will ever encounter.
The map has a history worth knowing. For much of the twentieth century, the core decision for most investors came down to two buckets: stocks and bonds. That binary came from the institutional world of pension funds and insurance companies, where the core decision was how much growth to chase and how much safety to buy. The framework spread into retail investing largely unchanged. Property was always part of the picture for big institutions, but it is only in recent decades that property, commodities, and then alternatives became standard formal categories — partly because academic research started measuring their behaviour separately, and partly because new financial products made them accessible to ordinary investors for the first time.
PIMCO, the fixed-income investment manager, captures the older logic cleanly. They frame bonds and cash as "defensive" asset classes that provide regular income, and stocks and property as "growth" assets that aim for long-term capital appreciation. That growth-versus-defensive split is useful, but it is a simplification. The deeper question is not whether something grows or defends — it is what mechanism drives the return at all.
Here is the most charitable version of the argument against building a map like this. Markets are complex and constantly evolving. The categories we use today were invented by people trying to describe a world that looked different from ours. Crypto did not exist. Infrastructure was a government function. Private credit was barely a retail category. Someone could reasonably argue that forcing every asset into a handful of classes is like insisting every living thing fits into the five kingdoms of an old biology textbook — technically tidy, practically limiting. That argument deserves to be taken seriously, and we will return to it. But here is why the map still holds: the categories are not defined by what things are called. They are defined by how they earn money. And there are genuinely only a small number of ways that can happen.
So let's build the map.
The first asset class is equities — shares. When you buy a share, you are buying a fractional ownership stake in a company. That ownership entitles you to a portion of whatever the company earns, either paid out as dividends or reflected in the rising price of the share as the company grows. So equities earn in two ways: capital gains and dividends. The risk that comes with this is the risk of the company — or the broader economy — doing worse than expected. You are not owed anything. There is no contract. You own a claim on future profits, and future profits are uncertain.
The second class is fixed income — bonds. A bond is a loan. You give money to a government, a municipality, or a company, and they promise to pay you interest at regular intervals and return your principal at a set date. J.P. Morgan describes the core examples: government bonds, municipal bonds, corporate bonds. What makes a government bond and a corporate bond the same asset class is not that they come from the same kind of issuer. It is that both earn their return through interest payments and principal repayment. The mechanism is identical. The risk differs — a corporate borrower can default in ways a government rarely does — but the return mechanism is the same loan structure. That is why they share a class.
On trading floors, fixed income is often bundled with currencies and commodities under the label FICC, which gives you a sense of how the edges of these categories can blur at an institutional level. But the core of the class is clear: you are a lender, and you earn because someone owes you money.
The third class is property — real estate. This one earns in two ways. Rent provides ongoing income; price appreciation provides a capital gain over time. PIMCO treats property as a growth asset alongside stocks, and for good reason — its return is tied to the economic activity of the people and businesses using the space. The risk profile sits between equities and bonds: more predictable than shares because leases create contracted income, but exposed to economic downturns that reduce demand for space.
The fourth class is cash and cash equivalents. This is the simplest mechanism: you deposit money with a bank or hold a short-term government instrument, and you earn a small, nearly certain yield. PIMCO describes cash as carrying the least risk of any class, and also the lowest return. The trade-off is explicit — you are not going to lose the nominal value, but you are not going to grow it much either.
One nuance worth carrying: Wikipedia flags that money market instruments are often labelled "cash equivalents" when they technically carry default risk, because they are short-term fixed-income instruments. Most retail-facing descriptions skip this distinction, but it matters if you want the map to be accurate. When a product is called a cash equivalent, it is worth checking whether it actually is one, or whether it belongs more accurately with fixed income.
The fifth class is commodities. Gold, oil, agricultural products, metals. These earn their return differently from everything else on the list. There are no dividends, no interest payments, no rent. The price moves in response to supply and demand in the physical world. Gold in particular has often had a low correlation with shares during bear markets — it has frequently held its value when shares were falling, though not every time — because its price is not driven by company profitability. PIMCO adds that commodities and real estate tend to outperform in high-inflation environments where traditional bonds underperform, because their prices are tied to the physical economy rather than fixed nominal payments.
Now pause, because this is worth restating before moving on. Five mechanisms. Owning a share of profits — that is equities. Lending money for interest — that is fixed income. Collecting rent and price appreciation — that is property. Holding something for a small but certain yield — that is cash. Holding something whose price moves with the physical world — that is commodities. Each one describes a distinct economic relationship between the investor and whatever the investment is doing in the real world.
The class is defined by that relationship. Not by the name. Not by the legal structure. Not by how you access it.
Which brings us to the second major idea. ETFs, managed funds, and superannuation accounts are not asset classes. They are containers.
ETFs and mutual funds are vehicles through which you access asset classes. They do not have their own return mechanism. An ETF that holds shares gives you equity risk. An ETF that holds bonds gives you fixed-income risk. The wrapper — the legal structure, the pricing mechanism, the fee arrangement — is the same in both cases. The underlying asset class is completely different. And the risk is completely different.
The practical difference between an ETF and a managed fund is about structure, not substance. ETFs trade throughout the day at a market price, and most are designed to track an index passively. Managed funds, often called mutual funds, price once a day at their net asset value, and many are actively managed. That distinction affects your costs and your access. It does not change what the underlying assets are or how they earn a return.
Superannuation in Australia is a particularly important example of this wrapper confusion. When Australians talk about "their super," they are usually talking about the account — the legal structure that holds the investments with particular tax treatment. But what matters for long-term outcomes is what that account actually holds. A balanced super fund might hold equities, bonds, property, cash, and infrastructure in varying proportions. The account itself does not earn a return. The asset classes inside it do.
Peel back any wrapper, and you will find one or more of the core classes underneath. The wrapper changes your costs, your tax treatment, your liquidity, and sometimes your access to things that would otherwise be hard to buy. It does not change your underlying economic exposure.
Now to what that means in practice, because the differences between asset classes are not just qualitative. Over long periods and across most developed markets, the ordering has been consistent: shares have delivered the highest returns and the largest swings, bonds lower returns with much smaller swings, and cash the lowest return with almost no swings at all. Private equity is usually projected to earn a little more than listed shares, with a rougher ride than its smoothed, infrequent valuations make it look. The exact numbers depend heavily on the period and the market chosen, which is why this series treats any single figure as an illustration rather than a promise.
Which leads to the third key idea: alternatives. And this is where the map gets genuinely contested.
Private equity, infrastructure, hedge funds, digital assets — these are grouped under "alternatives" in most financial writing, but that label is not a definition. It is a residual category for things that did not fit cleanly into the first five. PIMCO describes alternatives as "a broad range of investments" with no single definition, which is an honest acknowledgement that the category is a catch-all.
Infrastructure is the most interesting test case. Roads, airports, utilities, ports — these generate long-term contracted cash flows, often linked to inflation. The academic literature, according to the infrastructure asset class study, is still actively debating whether infrastructure constitutes its own distinct class or whether it is better understood as a bundle of equity and bond exposures dressed up in a different structure. One Australian study even found direct property delivering a better risk-adjusted return than infrastructure over its sample period — a result that complicates the idea of infrastructure as a superior alternative class. That debate is not resolved. The status of infrastructure as a separate class remains genuinely open.
Crypto is the harder case. Wikipedia lists cryptocurrency as a standalone asset class alongside commodities. J.P. Morgan groups digital coins with carbon credits under "alternative investments." Neither description explains the return mechanism. And that is the problem. The framework built here asks: what economic relationship does this investment represent? With equities, you own a claim on profits. With bonds, you are a lender. With commodities, you are holding something whose price tracks the physical world. With cryptocurrency, the return mechanism is not settled. It is not a claim on earnings. It does not pay interest. Its price does not track a physical commodity with industrial uses in any consistent way. Whether it belongs with commodities as a speculative asset, whether it has its own new mechanism, or whether it sits outside the map entirely — the sources drawn on here do not resolve that question, because it genuinely has not been resolved.
The steelman argument for treating alternatives as genuinely new classes is this: the map was built from twentieth-century markets. New instruments create new economic relationships. Infrastructure creates a relationship between the investor and essential public services over fifty-year time horizons. That is structurally different from buying a share in a listed company or lending money to a government. Dismissing it as "just equity and bonds in a trench coat" might be underestimating how much the economic relationship has actually changed. That argument has force.
But the honest response is that a new economic relationship only justifies a new class if that relationship is distinct, stable, and identifiable. Infrastructure is still being tested against that standard. Crypto has not yet met it by any consensus measure.
So here is where the map lands. The organising principle throughout has been a single question: how does this thing earn a return? Equities earn through ownership of profits. Bonds earn through lending. Property earns through rent and appreciation. Cash earns a certain but small yield. Commodities earn through price movements driven by physical supply and demand. Those five mechanisms cover almost everything a retail investor in Australia will encounter.
ETFs, managed funds, and superannuation accounts are legal structures that hold these classes. They shape costs and access. They do not change what is underneath. When you hold a balanced super fund, you hold a mix of asset classes — and understanding what those classes are, and what return mechanism each one uses, is more useful than knowing the name of the fund.
The ordering anchors the abstract map to something real: shares with the highest long-run returns and the biggest swings, bonds smoother and lower, cash lowest and steadiest, and private equity promising a little more than shares for a rougher ride than its reported valuations suggest.
The edges of the map are genuinely contested. Infrastructure may or may not be its own class — the academic literature has not settled this. Crypto's return mechanism has not been defined in a way that commands consensus. Treating those as open questions is not a weakness of the framework. It is an accurate description of where the field actually stands.
What the map gives you is a way of asking one clarifying question every time you encounter a financial product: what is the economic relationship here, and how does that relationship earn a return? Everything investable either has a clear answer to that question — and fits somewhere on the map — or it does not yet, in which case you know you are in contested territory.