A house view is a lens. It is the standing way a publication reads the world — what is driving prices right now, which forces matter more than the headlines suggest, and what it would take to be wrong about that. Point it at a jobs report, an earnings release, a closed shipping lane, and it tells you which of those touches your portfolio and which is noise.
Most readers already have one. It just lives in scattered places: a few beliefs about rates, a hunch about where the AI money is going, a rule of thumb about what inflation does to stocks. The trouble with a scattered view is that it cannot be checked. When a belief turns out to be wrong, nothing forces the question, and the belief survives by being forgotten.
This page is the fix for that. It is Grant Bishop's house view, written down, dated, and kept in one place. Everything else the publication produces hangs off it.
What it does for you
Markets throw off more information in a week than anyone can weigh. A house view is how you decide what to weigh. It answers three questions, in order:
What is going on? Not today's move — the current underneath it. Right now that current has two parts, and they pull against each other. The slow one: governments owe more than they can comfortably repay, and the path of least resistance is inflation run a little hot for a long time. The fast one: a war with Iran has kept the Strait of Hormuz mostly shut for six months, and the inflation that produces is the supply-shock kind — prices rising because goods cannot arrive, not because there is too much money chasing them.
13
What does that mean for a portfolio? Here the two currents disagree, and the disagreement is the useful part. Inflation of the first kind is friendly to stocks: companies raise prices with everything else, and earnings keep up. Inflation of the second kind is not: input costs rise faster than they can be passed on, and the 1970s are the record. So the old comfort — own good businesses and you will keep pace — is conditional. In a supply shock it holds for the businesses that are the supply, the ones whose product is the scarce thing. The rest are exposed. That single sentence is the frame, and it is what every piece here is measured against.
What would change that? This is the part most commentary leaves out, and it is the part that makes a view trustworthy rather than merely confident. A lens that cannot be wrong is not a lens. Every claim on the House View page names the evidence that would break it and the date it will be checked.
How it is built
Under the frame are five variables — the gauges the lens reads.
The market is paying less for good news: companies that beat and raise are being marked down anyway, which is a story about the cost of money, not about the companies.
67 A Fed chair who arrived with a mandate to cut is not cutting; his July meeting held rates with three members voting to raise.
9 Computing power for artificial intelligence has split into two businesses — capacity rented out by companies, and capacity funded by governments — and the two are now moving separately.
1315 Whether oil windfalls are funding the government side is an open question the page holds as a candidate. And grid power, not chips, is the constraint on how fast any of it gets built.
16
Each variable carries its evidence with a numbered source. Each will move, and when one does, the page moves with it.
How it stays honest
Below the variables sit the watchpoints. Each is a claim made now, in writing, with a date, the evidence that would prove it wrong, and the day it will be checked. One as it stands today:
W-5. The Warsh Fed does not cut this year, and a hike before
year-end is the base case. Wrong if: the 16–17 September meeting
cuts, or its statement signals an easing path. Check: FOMC
statement and projections. Review 30 Sep 2026.
911
On 30 September that row is in one of three states: it held, it tripped, or the evidence was mixed and the page says so. If it tripped, the row stays. It gets a date and a note in the changelog saying what was wrong and what replaced it. Nothing on this page is quietly edited. A view that revises its own history without a record is not a view; it is a sales pitch.
This is a different kind of accountability from a model portfolio with a return next to it. A return is mostly about entry price, sizing, and luck, and says little about whether the reasoning was sound. The record kept here is simpler and harder to fake: did the page say in advance what would change its mind, did it change when that evidence arrived, and is the change on the record. A revised view with a candid changelog entry is the most credible thing this publication will ever publish, and it will publish several.
What this page is not
It is not advice. Grant Bishop is a publication, not an adviser, and it operates deliberately inside the line the law draws between the two: general commentary, general circulation, no recommendation to anyone in particular. The decision, and the downside, stay with you. What you get is the reasoning, laid out so you can check it, and a standing promise that when it changes you will know.
It is not a forecast. A forecast is a number and a date. A house view is a claim about how the machine works — what is driving what, and which gauge to watch. If it is right, it stays right across a range of numbers. If it is wrong, it is wrong in a way you can name.
And it is not a person's opinion. Everything here is machine-drafted against a brief of primary sources, checked figure by figure back to the filing by machine, and read and committed by the publisher before it ships. That is stated on every piece because it is true, and because the alternative — implying a research staff the publication does not have — is exactly the kind of claim this page exists to refuse. The machines draft and check. The publisher decides what runs, names in advance what would prove the page wrong, and is there on the review date.
How to use it
Read the frame once. Then, when news arrives, ask which of the five gauges it moves. Most news moves none of them. Some moves one. Occasionally something arrives that a watchpoint named in advance, and that is the day the page changes — with the date on it.
What would change our mind
About the frame: two consecutive quarters of core inflation above 3 percent after oil has fallen back below $85 would say inflation has become the demand kind, and the old hedge is back. Broad-market valuations expanding while oil holds above $100 would say the market disagrees with the 1970s, and the page would have to find out why. Those are W-1's terms.
4
About the method: nothing. The method is the page. If watchpoints trip and the changelog fills, it is working.
Next check: the September Fed meeting. The row is on the Watch.