← Pieter Slegers hub  ·  Research hub  ·  Research library

Pieter Slegers — ETF Portfolio Update August 2026

Quality defined as three numbers (ROE, leverage, accruals), SPHQ as the month's spotlight, and a wide-moat add on both sides of the US/non-US split — MOAT at $114.1 and GOAT for €500 at €32.8.
2026-AUG-20 · Compounding Quality (Substack, paid post) · guest-authored by TJ Terwilliger, signed off "Team Compounding Quality" · written post · read ↗ · transcript · actionable insights
One-line take: the monthly ETF letter, and the clearest statement yet of what "quality" means when it has to be written as a rule a computer can run. The argument opens on Bessembinder — "the average cumulative return was +22,840%; the median cumulative return was −7.41%" — to make the case that a market-cap index buys the capital-destroyers along with the compounders. The month's spotlight, SPHQ (0.21% TER, physical, European share class SPQA), is the operational answer: score all 500 S&P constituents on exactly three inputs — return on equity, financial leverage and the accruals ratio — keep the top 100, and weight by quality score × market cap. Each input is argued from a chart: highest-ROE companies produce the highest returns, low debt does not help but high debt badly hurts, and companies with negative cash flow and negative earnings deliver negative returns. Two things are worth noting against the "quality" label. First, the disclosed top ten is 42.5% Information Technology and led by Lam Research at 5.51%, with Sandisk at 4.17% — memory and semicap equipment, i.e. exactly the part of the market the house side of this archive warned against on cycle grounds in July. Second, the July issue's own ETF of the Month, VFLO, does not appear anywhere in this month's portfolio charts. The transaction is a single wide-moat add on both sides: MOAT ($114.1) for Americans, GOAT (€500 at €32.8) for non-Americans, executed "tomorrow at the market opening."

1. Stocks & names mentioned

Five funds carry a view. As in the July issue, the portfolio's other ETFs are named in the charts but their tickers are not printed, so they are transcribed in transcript.txt and not given rows — nothing is inferred. The ten SPHQ constituents disclosed in the holdings table (Lam Research, Visa, Mastercard, GE Vernova, Apple, Sandisk, General Electric, Costco, KLA, Cisco) are index composition, not picks, and likewise get no rows. Research legend: SA Seeking Alpha · STK Stock Analysis (funds carry no Qualtrim page). Written post with no timestamps — the At link opens the article.

TickerNameResearchViewWhat he saidAt
SPHQInvesco S&P 500 Quality ETFSA · STKPositiveETF of the Month. Disclosed specification: TER 0.21%, physical replication, ISIN US46137V2410; European share class SPQA (ISIN IE000E6TPCH9). Method: start from the S&P 500, score every constituent on ROE, financial leverage and the accruals ratio, "keep only the 100 companies with the highest scores… weight each company based on its quality score and market capitalization." The case for each input is argued from data — highest-ROE quintile earns the highest returns (MSCI); low leverage doesn't help returns but high leverage hurts them (MSCI); companies with negative cash flow and negative earnings deliver negative returns (AQR). Sector mix: Information Technology 42.5%, Industrials 18.9%, Financials 15.3%. Top ten led by Lam Research 5.51%, Visa 4.90%, Mastercard 4.85%, GE Vernova 4.60%, Apple 4.58%, Sandisk 4.17%.read ↗
MOATVanEck Morningstar Wide Moat ETFSA · STKPositiveBOUGHT — the month's add for the American portfolio, at a stated price of $114.1, "tomorrow at the market opening." The reasoning is one line — "Let's add some quality, Wide Moat stocks this month!" — with no valuation, screen or trigger given. Already the portfolio's third-largest holding at 15.0% and its third-best performer at roughly +34%, so this is an add to a winner, not a new position.read ↗
GOAT.ASVanEck Morningstar Global Wide Moat UCITS ETFSTKPositiveBOUGHT — the non-American twin of the MOAT add: €500 at a stated price of €32.8, same "tomorrow at the market opening" timing. Held at 15.0% of the non-American book, where it is the worst of the paired positions — roughly +21% against MOAT's +34% on the American side, the widest gap between the two portfolios in the comparison chart. The post does not comment on that divergence.read ↗
USMViShares MSCI USA Min Vol Factor ETFSA · STKNeutralHeld at 12.1% of the American portfolio and its weakest position at roughly +9%, less than a quarter of the best. Bought only four weeks earlier at $98.35 on a "widest low-volatility spread since 2006" argument; no comment is offered on it here, and no add is made to it this month.read ↗
MVOL.LiShares Edge MSCI World Minimum Volatility UCITS ETFSTKNeutralThe non-American min-vol sleeve, at 14.6% of that portfolio and its weakest position at roughly +14% — the same last-place ranking USMV holds on the American side, though a few points better. Bought on 30 July at €76.91; not added to and not discussed here.read ↗

Three things a reader should carry away from the charts rather than the prose. (1) VFLO is gone. The July ETF of the Month — bought and written up four weeks ago — appears in none of this issue's weight or performance charts, and the disappearance is not mentioned. (2) The quality label and the holdings pull in different directions. SPHQ's disclosed top ten is led by Lam Research (5.51%) and includes Sandisk (4.17%): memory and semicap names whose earnings are cyclical, in a fund selected on ROE, leverage and accruals — precisely the trap the house side of this archive flagged when it refused the memory complex in July because "these companies don't have any pricing power." A trailing-ROE screen buys a cyclical at the top of its cycle by construction. (3) The two portfolios are not twins. The paired comparison shows the non-American multifactor sleeve at roughly +62% against +28.5% for the American one, and the wide-moat pairing inverted (+21% vs +34%) — differences the post waves away with "over time, I expect both portfolios to generate similar returns."

2. Talking points

The Bessembinder opening

Quality, input one: high reinvestment

Quality, input two: low leverage

Quality, input three: real earnings (the accruals ratio)

SPHQ: the whole method in four steps

The four factor sleeves

Two portfolios, because of a rule rather than a view

The American book

The non-American book

The transaction

What is not said

3. In plain English

A jargon-free summary of the thesis behind each argued fund. (Renders on each name's consolidated page.)

SPHQ — Invesco S&P 500 Quality ETF Positive

An ordinary S&P 500 fund buys all five hundred companies in proportion to their size, which means it buys the excellent ones and the value-destroying ones alike. The argument for doing something else is a statistic from Hendrik Bessembinder: across nearly a hundred years, the average American stock returned about 22,840% while the median one lost 7.4%. Almost all the gains come from a handful of names; most stocks are dead weight.

This fund tries to keep only the good part by scoring all five hundred companies on three measurable things and keeping the best hundred. The three are: how much profit a company earns on the money shareholders have put in (return on equity); how much it has borrowed (leverage); and how closely its reported profits match the actual cash arriving in the bank (the accruals ratio — a low number means the profits are real rather than accounting judgement). It then weights each survivor by its score and its size. The fee is 0.21% a year, and there is a European version, SPQA, for readers who cannot buy US funds.

One caution the letter does not raise. Because all three tests look backwards, a company earns its way into the fund when its recent profits are at their best — which is exactly the top of the cycle for a cyclical business. The disclosed top ten is more than 40% technology and led by Lam Research and Sandisk, both semiconductor names whose profits swing hard with the memory cycle. The same archive's stock-picking side refused that complex a month earlier on precisely those grounds.

MOAT — VanEck Morningstar Wide Moat ETF Positive

This fund holds American companies that Morningstar's analysts judge to have a "wide moat" — a durable structural advantage that should let them keep earning high returns for a couple of decades — and, importantly, buys them only when Morningstar also thinks they are cheap relative to its own estimate of fair value. So it is a quality screen with a valuation step attached, which makes it an unusually close match to the way the rest of this archive picks individual shares.

The August purchase is a top-up: money goes in at the next market open, at a quoted $114.1, into a position that is already 15% of the American portfolio and its third-best performer at roughly +34%. No price target, screen or trigger is offered — the whole justification is one sentence about adding wide-moat exposure this month, which is a materially thinner standard than the equity book's named limit prices.

GOAT.AS — VanEck Morningstar Global Wide Moat UCITS ETF Positive

The same idea as MOAT, but worldwide and in a European wrapper, because readers outside the United States are not permitted to buy US-domiciled funds (and Americans cannot buy this one). €500 goes in at the next open, at a quoted €32.8.

The detail worth noticing is in the comparison chart rather than the text. Of all the paired positions, this is where the two portfolios diverge most in the American book's favour: the global version has returned roughly 21% against roughly 34% for the US-only version. A global moat basket has been the weaker way to own the same idea, and the letter's only comment on any such gap is a general expectation that "over time… both portfolios [will] generate similar returns."

USMV — iShares MSCI USA Min Vol Factor ETF Neutral

A fund that holds the least jumpy large American shares. It was bought only four weeks earlier, at $98.35, on the argument that the gap between calm and volatile stocks was the widest since 2006 — the kind of trade you make expecting the gap to close.

A month on it is the weakest of the seven American positions, at roughly +9% against roughly +42% for the best, and it gets no mention at all in this issue: no add, no review, no comment on whether the spread that justified it has moved. Worth watching, because a position bought on a specific one-off argument and then left undiscussed is the kind that quietly stops being monitored.

MVOL.L — iShares Edge MSCI World Minimum Volatility UCITS ETF Neutral

The non-US twin of USMV: the calmest shares from around the developed world, in a European wrapper, bought on 30 July at €76.91 on the same "widest spread since 2006" reasoning.

It occupies 14.6% of the non-American portfolio and, like its American counterpart, sits last on performance at roughly +14%. Neither is added to this month, and neither is discussed — the whole transaction budget goes into the wide-moat sleeve instead.


Summary derived from the archived Compounding Quality post (text in transcript.txt) for personal study. Not investment advice. © Compounding Quality / TJ Terwilliger for source material.