Howard Marks, in The Most Important Thing, introduces a concept that separates adequate investors from exceptional ones: the difference between first-level and second-level thinking.
First-level thinking says: "This is a good company. I should buy the stock."
Second-level thinking says: "This is a good company, but everyone already knows it's a good company. The stock is priced to perfection. What would need to be true for it to outperform from here — and is that likely?"
Applied to Tokyo real estate, the distinction is sharp, practical, and often financially consequential.
First-Level Thinking in Tokyo Real Estate
First-level thinking about the Tokyo market sounds like this:
- "Tokyo is a global city with a shrinking population nationally but growing population in the core — therefore Tokyo real estate is a good investment."
- "Shinjuku has excellent transport connectivity, high rental demand, and strong liquidity — therefore a property in Shinjuku is a good investment."
- "Japan has low interest rates and stable rent — therefore leverage is cheap and returns are reliable."
All of these statements are true. They are also widely known, frequently repeated, and already priced into the market. First-level thinking produces consensus conclusions. Consensus conclusions, in efficient markets, generate consensus returns — which means average or slightly below average after fees and costs.
The problem with buying a property in Shinjuku at 240x GRM because "Shinjuku is excellent" is not that the reasoning is wrong. It is that you are paying a price that already reflects everything you know about why Shinjuku is excellent. To outperform from that price, something would need to improve further — and it already has.
What Second-Level Thinking Looks Like in Tokyo
Second-level thinking about Tokyo real estate is slower, more uncomfortable, and more contrarian. It asks different questions:
On location: Not "is this a good station?" but "is this station priced as if it is good, priced as if it is great, or priced as if it is declining — and which of those reflects reality?"
At some stations, the data shows a GRM 20–30% above the historical median for that ward. That premium reflects market consensus about the station's quality. Second-level thinking asks whether the consensus is correct — or whether it has overshot.
At other stations, the GRM sits below the ward median without an obvious structural reason. First-level thinkers avoid them because they "don't feel right". Second-level thinkers investigate: is there a data story that contradicts the sentiment?
On timing: Not "is now a good time to buy in Japan?" but "what is the market pricing in about the next 5–10 years — and do I agree with that implicit forecast?" Our analysis of Tokyo macro drivers covers the key variables behind this question.
When GRM at core Tokyo stations is at a multi-year high relative to historical medians, the market is pricing in continued rent growth, continued low vacancy, and continued demand from international capital. That may be correct. Second-level thinking forces you to specify the conditions under which it would not be — and how likely those conditions are.
On "safe" assets: Not "this building is new, the developer is reputable, the location is prime" but "what am I paying for that safety premium — and is there a comparable risk-adjusted return available in a less obvious place?"
New buildings in branded locations carry significant goodwill premiums. The depreciation curve on a new condominium is steepest in the first 10 years. Second-level thinking recognises that the safest-seeming entry point is often the most expensive one on a risk-adjusted basis.
The Station Median Divergence Test
One practical application of second-level thinking is tracking the deviation between a listing's GRM and the current station median. Our station-level methodology explains how we calculate these medians across Tokyo's residential market.
If a listing is priced at GRM 180x and the station median is 220x, the market is implicitly saying: "something is wrong with this listing — it should be cheaper." Second-level thinking asks: is the market right, or is this a mispricing?
Often the market is right. The listing may have a hidden defect (management problems, adverse building conditions, a tenant locked into a below-market rent). The discount is earned.
But sometimes the market is reflexively avoiding a station due to name recognition or vague sentiment, while the underlying transaction data — rent levels, vacancy, transaction volume — tells a different story. That gap is where second-level thinking generates returns.
Tokyo Insights tracks this gap systematically: the deviation between listing price and station-level median, adjusted for layout and age. It is not a sufficient condition for a good investment. It is a starting point for the right questions.
Second-Level Thinking on the Macro
The most commonly repeated macro thesis for Japan is: "Interest rates are low, the yen is weak for foreign buyers, and institutional capital is flowing in — therefore it is a good time to buy."
Second-level thinking deconstructs this:
- Interest rates are low — but mortgage rates for non-residents are not as low as domestic rates (typically 2–4% vs 0.5–1.5%). The spread matters enormously for Cash-on-Cash returns.
- The yen is weak — this makes entry cheap in foreign currency terms, which is real. But it also means exit proceeds are worth less in home currency if the yen strengthens. A 15% yen appreciation reverses a large part of a 5-year yield advantage.
- Institutional capital is flowing in — which means professional competition is higher than it was, and the easy mispricings have been partially closed. The opportunity is more selective than five years ago.
None of this means Japan is a bad investment. It means the easy consensus trade — "Japan is cheap and safe" — is less obviously correct than it was when fewer people believed it.
The Practical Implication
Second-level thinking in Tokyo real estate does not mean being contrarian for its own sake. It means asking — before every decision — whether the price reflects only what everyone else can see, or whether there is a non-consensus element the data supports.
The stations worth analysing seriously are not always the ones in the headlines. The right GRM range is not the one that "feels safe". The right entry point is not when market sentiment is most positive.
Most of this sounds obvious stated directly. The difficulty is that first-level thinking feels more comfortable in the moment — it comes with consensus validation. Second-level thinking requires the discipline to sit with discomfort and follow the data even when it leads somewhere unfamiliar.
That is where the real Tokyo real estate opportunity has always been.
Tokyo Insights tracks GRM deviation from station medians across Tokyo's residential market. If you want to know whether a specific listing sits above or below market consensus — and what the data says about that gap — submit it for independent analysis.
Related guides
- What is a good GRM in Tokyo? Station-by-station benchmarks
- Ward vs station: why the right unit of analysis changes everything
- Tokyo macro drivers 2026: what the data says about the market
- Tokyo deal screening playbook: how to filter listings systematically
- The barbell strategy for Japanese real estate: Taleb's framework applied