Personal Finance (1): Why Asset Allocation Matters

Until six months ago my entire financial strategy was: salary in, rent out, the rest sits in a checking account at 0.2%. I write distributed systems for a living, but I had never once thought seriously about where my money should live. I’m not a CFA and this isn’t a finance blog — it’s an engineer writing down a framework as he builds it, because I know a lot of engineers with the same blind spot.

The Slow Robbery Nobody Talks About#

China’s CPI inflation has averaged about 2% over the past decade. In engineering you’d round 2% down to “basically zero.” On your money, it isn’t.

The math that woke me up. Money kept in cash grows at its nominal rate, but what it can buy grows at roughly (nominal rate − inflation). Leave 100,000 RMB in a 0.2% account for 10 years:

• Nominal: 100,000 × 1.002¹⁰ ≈ 102,000 RMB (the number on screen goes up)
• Real: 102,000 ÷ 1.02¹⁰ ≈ 83,700 RMB in today’s purchasing power

You lost over 16,000 RMB in real value — no crash, no scam, no bad bet. You lost it by doing nothing. In engineering, “don’t touch anything” is safe. Here it isn’t: inflation nibbles at cash every year you leave it still. That arithmetic is what pushed me to learn allocation. If standing still costs 16% of my purchasing power per decade, I have to do something.

The Egg Basket Problem Is Deeper Than the Proverb#

“Don’t put all your eggs in one basket” is the first thing anyone says about investing. My engineering brain pushed back: if you find the strongest basket — a top fund, a blue chip — why not go all in? Three market crashes changed my mind.

The pattern: different asset classes fail at different times, for different reasons. Stocks crash on economic panic; bonds dip when rates move; gold spikes when fear runs high. So the real point of the proverb isn’t “use three baskets” — it’s that the baskets shouldn’t be made of the same material. Three stock funds all fall apart in the same storm. Real diversification means owning things that react differently to the same event: stocks and bonds, domestic and international, financial and physical. In the 2015 case, a 50/50 stock-bond split turned a −45% wipeout into about −22% — the difference between “recover in a few years” and “never recover psychologically.”

Risk and Return: No Free Lunch#

I wasted months looking for the “best” investment: high return, low risk. It doesn’t exist — the same way CAP theorem says you can’t have consistency, availability, and partition tolerance at once. You can’t have high returns and low risk simultaneously. Driving is the analogy I keep: 30 km/h means you arrive slowly but can stop for a kid in the road; 200 km/h means a lap record or you don’t arrive at all. Investments spread along the same spectrum:

“Average” is the trap for index funds: nobody earns 10% in a given year. You get +25%, then −15%, then +32%, and it averages out. The one rule that took seeing the table to click: there is no highway speed at bicycle risk. Anyone selling one is lying, confused, or about to lose money. And in today’s low-rate China, the bottom of this table has sunk right up against inflation: as of mid-2026 a checking account pays ~0.2%, Yu’E Bao’s 7-day yield fell below 1% for the first time on record (~0.88%), a 1-year bank deposit is ~1.3%, bank wealth-management averages ~1.8%, and even the 10-year government bond yields only ~1.74% — nearly all of it sitting at or under ~2% inflation, meaning “safe” money quietly loses purchasing power right now. So the right question isn’t “which investment is best?” It’s: how fast do I need to go, and how much turbulence can I stomach?

What “Allocation” Actually Means#

I thought allocation meant picking the right stocks or funds. It doesn’t. Allocation is about proportions — what percentage of your money goes into each asset class (stocks, bonds, cash, gold, real estate). Same ingredients, different recipe, different result:

Here’s the surprise. Brinson, Hood, and Beebower (1986, updated 1991) analyzed what drives portfolio returns and found that roughly 90% of the variation over time comes from the allocation decision, not from which specific funds you picked. All the noise — the stock tip from your cousin, agonizing over Fund A vs Fund B — is maybe 10% of the outcome. The other 90% was decided when you chose 60/30/10. As an engineer this stung: 90% of a system’s performance is architecture, only 10% is the specific code. Turns out finance works the same way.

Three Dimensions Decide Your Split#

Once allocation is about proportions, the question is: which proportions for me? Three things decide it.

1. What you can actually buy. In China a normal person can access: cash / money market (Yu’E Bao and clones), bank deposits and structured deposits, bond funds, stock index funds (CSI 300 = large-cap, CSI 500, ChiNext), gold (bars, paper gold, gold ETFs), and real estate (huge minimum, terrible liquidity). Article 2 unpacks each.

2. Time horizon — when you need the money. A 25-year-old investing for retirement has ~35 years; a −40% crash in year 3 is survivable because there are 32 years to recover, and historically the market always has. A 55-year-old retiring in 5 years faces the same crash as a disaster: no time to recover, possibly forced to sell at the bottom. Same salary, same net worth, same personality — completely different optimal split, purely from time. The mistake I nearly made was building a 5-year-runway portfolio when I had 30+ years, because conservative felt safer. Over decades, slow is expensive.

3. Risk tolerance — your nervous system, not just math. Some people watch a −30% drop and think “stocks are on sale.” Others (probably most of us, me included) feel a knot at −5% and check the app twelve times a day. Neither is wrong; they’re different wiring. It matters because risk tolerance sets the maximum volatility you can endure without panic-selling — and panic-selling at the bottom turns a temporary paper loss into a permanent real one. Be honest: if a −30% drawdown would make you sell everything at 3 AM, don’t put 80% in stocks. The math that says you’d win long-term assumes you hold through the dips. If you can’t hold, the math is fiction.

How to pick, concretely: take the lower of what your time horizon allows and what your stomach allows. Long horizon + steady nerves → aggressive. Short horizon or jumpy nerves → dial down the stock share until you’d hold through the worst year in the table above without selling.

The Enemy Is Doing Nothing#

Back to the opening, now with 200,000 RMB over 10 years and inflation at 2%.

The gap in real purchasing power: 126,000 RMB — a 63% difference, on the most boring allocation imaginable, the beige Honda Civic of strategies. No stock-picking, no market timing. Just not leaving everything in cash. Scenario B isn’t risk-free: in some of those years the stock half would be down 20-30% and you’d need the discipline not to sell. Bumpier journey, better destination. The most dangerous financial decision isn’t picking the wrong fund — it’s not making a decision at all.

What’s Coming Next#

This article was the why: why diversification is structural rather than a proverb, why the recipe matters more than the ingredients, why inaction is itself a costly choice.

Article 2 is the product zoo: every major asset class a regular person in China can buy — what a money market fund really is, how bond funds work under the hood, whether gold is an investment or a shiny rock — with real numbers and real trade-offs.#

This is Part 1 of the Personal Finance series. Next: Part 2 — The Product Zoo: From Money Market Funds to Gold .