DEV Community: Ruslan Averin

The 60-Day Clock Ran Out and Brent Went Through $91

Ruslan Averin — Tue, 18 Aug 2026 23:08:23 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

The deadline set at Versailles in June gave the parties sixty days to end the war with Iran and settle the nuclear question. It expired on 17 August. Nobody conceded, and nobody asked for more time.

Brent went above $90 and printed $91.36. West Texas Intermediate settled at $84.50, up 2.6% on the day the window closed.

What the price is actually saying

Instrument	Level
Brent, after deadline lapsed	$91.36
Brent, settle earlier in week	$90.87 (+2.7%)
WTI, settle	$84.50 (+2.6%)
Strait of Hormuz	still closed
US blockade	still in place

A geopolitical risk premium normally decays. The event happens, the supply disruption turns out smaller than feared, traders unwind, and within a few weeks the curve looks like it did before.

That is not what this is. The premium here is not pricing the risk of a disruption — the disruption already exists and has a name. The strait is shut, the blockade is running, and the deadline that was supposed to end both came and went without moving either.

The part that matters more than the barrel price

Oil at $91 is uncomfortable. Oil at $91 with no mechanism visible for it to come down is a different input to every model that has a rate path in it.

That is why the more revealing number this week was not in the commodity market at all. The 30-year Treasury yield reached 5.311%, its highest since June 2007, and the reason cited was oil feeding inflation expectations at the same time as the government's borrowing needs keep growing.

Energy is the transmission belt. It runs from a stalemate in the Gulf, through headline inflation, into the long end of the curve, and from there into the discount rate applied to every equity on the exchange. The chip selloff and the 30-year auction are not separate stories from this one.

What breaks the picture

Three things, in descending order of likelihood.

A partial reopening. Not a deal — a technical arrangement that lets a share of tanker traffic through. That alone would take several dollars off Brent, because the market is pricing zero.

Demand destruction. Sustained prices in the nineties do their own work on consumption, with a lag of quarters rather than weeks. This is the slow route and it arrives whether anyone negotiates or not.

A genuine agreement. Possible, and the pricing implies the market rates it low after watching sixty days produce nothing.

How I read it

I would separate the trade from the narrative. The narrative is a diplomatic failure, and diplomatic failures are hard to time.

The trade is simpler: as long as the strait is closed, the marginal barrel is expensive, and everything that discounts future cash flows has to work against a higher rate. That is an argument for owning the producers and the tanker operators rather than betting on the resolution — and for treating any equity valuation built on a 2025 discount rate with suspicion.

The single number I would watch next is not Brent. It is the 30-year. If the yield keeps making new highs on oil, the equity market has not finished repricing.

More market analysis by Ruslan Averin at averin.com.

Palo Alto Hit a Record at 85 Times Earnings. The ARR Explains Half of It

Ruslan Averin — Tue, 18 Aug 2026 23:07:40 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Palo Alto Networks reached a record high at roughly 85 times earnings. Next-generation security annual recurring revenue hit $8.13 billion, up 60%.

Then the detail that most of the coverage put in a subordinate clause: about $1.6 billion of that ARR came from the CyberArk and Chronosphere acquisitions.

The numbers, separated

Metric	Value
NGS ARR	$8.13B
Growth	+60%
Of which acquired	~$1.6B
Implied organic ARR	~$6.5B
Multiple at record	~85x earnings
FY2026 EPS guidance	$3.77–$3.79, raised
Stock, mid-June to mid-July	$287.78 → $358.68 (+24.6%)
Reaction on results day	−5.64%, then four weeks to a record

Back the acquisitions out and the growth rate is still good. It is not 60%.

Why the distinction is the whole argument

Organic recurring revenue compounds. A customer who adopts a platform and expands within it produces growth next year without further capital being spent.

Acquired recurring revenue is bought. It is real revenue with real customers, but it was paid for once with shareholder capital, and it does not repeat unless more acquisitions follow. Growth by acquisition has to be re-purchased every year, at prices set by whoever is selling.

A market paying 85 times earnings is paying for the first kind. The disclosure says a meaningful slice is the second kind. That gap is the entire risk in the position, and it does not appear anywhere on the price chart.

The pattern in the price is worth reading

The stock fell 5.64% on results day, then climbed for four straight weeks to a record.

That sequence usually means the first reaction was to the headline and the second was to the reading. Sell-side targets moved up substantially through those weeks — Piper Sandler to $345 from $265, Evercore holding Outperform at $415, Needham to $425, Tigress to $430 — and the stock followed the revisions rather than the print.

Both moves were rational. The initial disappointment was about the quality of the beat; the recovery was about the platform thesis, which says a security vendor consolidating categories under one contract earns durable pricing power and a higher multiple.

That thesis may well be right. It is also precisely what an 85x multiple already assumes.

The rate problem in the background

An 85x multiple is a long-duration asset by definition — the earnings justifying it are years away.

Those earnings are discounted at a rate anchored to the long end of the Treasury curve, which reached 5.311% this week, the highest since June 2007. The same arithmetic that took SMH down 9.5% in a month applies here, and it does not care that cybersecurity spending is defensive.

Palo Alto has so far been exempt, rising while semiconductors fell. Exemptions of that kind are usually temporary.

How I read it

The business is executing and the platform strategy is coherent. My reservation is entirely about what the price requires from here.

At 85 times earnings with a rising discount rate, the stock needs organic acceleration, not more acquisitions — because more acquisitions confirm exactly the concern that the $1.6 billion raises.

The disclosure I would want is organic ARR growth stated separately, quarter by quarter. Companies growing organically publish that number readily. Companies that would rather discuss the blended figure tend to have a reason.

More market analysis by Ruslan Averin at averin.com.

Bitcoin at $65,000 Is Down 30% and Nobody Is Calling It a Crash

Ruslan Averin — Tue, 18 Aug 2026 23:07:09 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Bitcoin trades near $65,000, a market capitalisation of about $1.3 trillion. It started the year around $93,000.

That is roughly a 30% drawdown, and it has produced almost no commentary. Ethereum is down 1.8% on the week, bitcoin down more than 3%, and the loudest debate in the space is about the timing of a bottom rather than about the size of what has already happened.

The state of the market

Metric	Level
Bitcoin	~$65,000
Market capitalisation	~$1.3T
Start of 2026	~$93,000
Drawdown	~30%
Support	$62,500
Resistance	$65,000–$70,000
Moving averages	price below key levels
Weekly projection	$64,330–$66,825

Why nobody called it a crash

Speed, not size, is what makes a market move into a story.

A 30% fall in a week is a crisis with a name and a cause. The same 30% spread across seven months is a repricing that people adjust to a few percent at a time, and by the end nobody remembers what the starting price was.

The arithmetic is identical. The narrative is completely different. This matters practically, because portfolios that would never tolerate a crash routinely sit through the slow version without acting.

What is actually pressing on the price

Bitcoin has no cash flows. Its price is a function of two things: how much liquidity exists, and how willing people are to hold risk without being paid to wait.

Both have moved against it this year, and the bond market is where you can see why.

With the 30-year Treasury at 5.311% — the highest since June 2007 — a risk-free instrument pays over five per cent for doing nothing. Every non-yielding asset competes against that, and the competition has become considerably harder over seven months. That is the same force compressing semiconductor multiples, and bitcoin, being the purest long-duration asset in existence, feels it most directly.

The uncomfortable implication for the "digital gold" thesis: gold's case strengthens when real yields fall. This year they have risen. Bitcoin has traded like the highest-beta technology asset on the board, not like a hedge — which is what it has done in every previous rate cycle too.

The level that matters

Analysts are watching $63,220 as a weekly close level, with $62,500 as range support. Below the range, the next thing supporting the price is not technical, it is whoever decides the drawdown is finally large enough.

Above, $70,000 is where the range gives way and the seven-month downtrend would need re-examining.

Neither level is magic. But in an asset with no earnings to anchor value, ranges are what the market has instead of valuation, and they are worth respecting for exactly that reason.

How I read it

I would stop treating bitcoin as an uncorrelated allocation. It has spent this year trading as a leveraged expression of the same liquidity conditions that set the Nasdaq's multiple, and the 30-year yield explains more of its path than any crypto-native development has.

That makes it a rates trade wearing different clothes. If the long end stabilises or falls, the case improves quickly and mechanically. While the 30-year keeps making nineteen-year highs, the opportunity cost of holding a non-yielding asset keeps rising, and no amount of adoption narrative offsets that arithmetic.

The bottom, when it comes, will most likely be visible in the bond market before it is visible on a crypto chart.

More market analysis by Ruslan Averin at averin.com.

Walmart Can Move $42 Billion of Its Own Value in One Session

Ruslan Averin — Tue, 18 Aug 2026 23:06:38 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Options pricing implies Walmart could move $42.1 billion of market value on Thursday's result.

That is not a typo, and it is worth sitting with. Forty-two billion dollars is more than the entire market capitalisation of most companies in the S&P 500, and it is the amount that could change hands in one session on a single retailer's quarterly disclosure.

The setup

Item	Value
Implied swing on earnings	$42.1B
Share price	~$116
Year to date	~+1%
July low	$106.79
Reports	Thursday 20 August
Last quarter	In line, guidance unchanged — and punished for it

Note the year-to-date figure. Up about 1%, in a year when Target ran 55%. Walmart has been dead money since a quarter that was merely in line with unchanged guidance, which tells you what this market does to a company that meets expectations without raising them.

What is actually being priced

Two pressures land on the same income statement, both coming from the same place.

The customer has less. Brent above $90 after the Iran deadline expired is a direct levy on discretionary spending. Every dollar at the pump is a dollar not in the basket, and it hits Walmart's core customer hardest because that customer has the least slack.

The company pays more. Walmart's own distribution runs on fuel. Elevated energy prices raise the cost of moving goods to shelves, and that cost lands on gross margin regardless of how well the stores execute.

Retail sales came in weaker than expected the Friday before this week. That is the backdrop for every retail print this week and it is why Schwab described the market as sitting flat with yields up, waiting for the retailers.

The thing Walmart tells you that nobody else does

Walmart sells to a cross-section of the country every week. That makes its management commentary the most valuable single read on consumer condition available anywhere — better than official data, because it arrives faster and it comes with observed behaviour rather than survey responses.

The specific things worth listening for:

Trade-down. Are higher-income households shopping at Walmart more? That is bullish for Walmart's market share and bearish for everything above it in the retail stack.

Basket size versus visit frequency. More frequent, smaller baskets means households are managing cash flow week to week. That is a stress signal regardless of what the revenue line says.

Category mix. Groceries holding while general merchandise softens is the classic shape of a consumer under pressure, and it changes what the whole retail complex is worth.

How I read it

The $42.1 billion implied move is the market admitting it does not know which way this goes — and that is unusual for a company as predictable as Walmart.

Priced at roughly 1% year-to-date gains, the stock is not carrying the optimism that Target's 55% is. That asymmetry matters: Walmart has been punished already, Target has not.

What I would watch is not the beat. It is whether management raises guidance or repeats it. This company was marked down once already this year for an in-line quarter with unchanged guidance. Doing that twice in a row, with fuel costs where they are, would tell you something about the consumer that no single quarter's numbers would.

More market analysis by Ruslan Averin at averin.com.

Target Is Up 55% This Year. Analysts Still Will Not Buy It

Ruslan Averin — Tue, 18 Aug 2026 23:06:07 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Target has gained roughly 55% this year. Wall Street's response is 12 Buys, 23 Holds and 3 Sells, with an average price target implying about 5% downside.

That is not a disagreement about the company. It is a disagreement about the price.

Two setups, one message

	Target	Lowe's
Consensus EPS	$2.35 (+13% y/y)	$4.22 (−3% y/y)
Consensus revenue	$26.15B (+3%)	$26.14B (+9%)
Comparable sales expected	~2.6%	softer
Stock, year to date	+55%	selloff into print
Ratings	12 Buy / 23 Hold / 3 Sell	24 Buy / 10 Hold / 1 Sell
Options-implied move	~7%	—

Read the two revenue-and-EPS pairs against each other and the whole retail quarter is in there.

Target: revenue up 3%, EPS up 13%. Profit growing faster than sales — margin recovery, mix improvement, the shape of a genuine turnaround.

Lowe's: revenue up 9%, EPS down 3%. Sales growing three times faster while profit shrinks. That is volume bought rather than earned, and the buying happens through price.

What Target actually has to prove

The first quarter was excellent: comparable sales up 5.6%, traffic up 4.4%, a 17% EPS beat, guidance raised. That quarter is what produced the 55%.

Consensus for this quarter is 2.6% comparable sales — less than half the first-quarter rate. So the question is not whether Target beats a lowered bar. It is whether the deceleration from 5.6% to 2.6% is the recovery maturing or the recovery ending.

Traffic answers that, not sales. Sales can be held up by ticket for a quarter or two while visits decline; traffic cannot be faked. If traffic growth holds near the 4.4% of Q1, the story is intact and the Holds are wrong. If traffic goes flat while sales hold, the 55% has nothing left to stand on.

The macro underneath both

Retail sales came in weaker than expected the Friday before this week, which is why these reports carry more weight than usual. Fuel costs are elevated with Brent above $90 — that is a direct tax on the consumer's discretionary budget and a direct cost on the retailer's distribution.

Both companies are being asked the same question at the same moment: when the customer has less to spend, what do they protect? Target's answer shows up in traffic. Lowe's shows up in the gap between its revenue line and its EPS line.

How I read it

I find the Lowe's setup more informative and the Target setup more dangerous.

Lowe's expectations are already low — a soft quarter is priced, and 24 Buys against 1 Sell means the sell-side sees the housing freeze as temporary. There is little asymmetry either way.

Target is the one carrying the risk. A 55% run into a quarter where growth is expected to halve, with the sell-side refusing to upgrade and the options market pricing a 7% move, is a setup where a good result gets a shrug and a mediocre one gets punished.

I would want to see the traffic line before the headline. It is the only number in either report that cannot be manufactured.

More market analysis by Ruslan Averin at averin.com.

Home Depot Beat, Held Guidance, and Called the Housing Market Frozen

Ruslan Averin — Tue, 18 Aug 2026 23:05:36 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Home Depot reported fiscal second-quarter sales of $47.9 billion, up 5.7%, beat on both the top and bottom lines, and reaffirmed its full-year outlook. Chief financial officer Richard McPhail then described the operating environment as "frozen housing market conditions."

Both statements are true at once, and the tension between them is the whole story.

The quarter

Metric	Q2 FY2026
Sales	$47.9B
Growth year on year	+5.7%
Guidance, full-year sales growth	2.5%–4.5%, reaffirmed
Guidance, full-year EPS growth	flat to 4.0%, reaffirmed
Management commentary	Results "exceeded our expectations"
Demand shape	"Smaller projects", broad based

Beat, no raise — and why the market shrugged

A company that beats and reaffirms is telling you the beat was not structural. If the quarter had changed the year, the year would have been changed.

Reaffirming after a beat means one of two things: management is being conservative, or the strength was in a category they do not expect to persist. Here it is closer to the second. Growth came from smaller projects — maintenance, repairs, incremental improvements to homes people already live in.

That is real revenue and it is defensible. It is also not the demand that Home Depot is built for.

What "frozen" actually means

The home improvement business runs on housing turnover. Someone buys a house, and within eighteen months they replace the kitchen, redo the bathrooms, refinish the floors. Those are large-basket, high-margin projects, and they are triggered by a transaction.

Transactions require mortgages. Mortgages track the 10-year Treasury, which sat near 4.72% this week while the 30-year reached 5.311%, its highest since June 2007.

At those rates, an owner with a mortgage from the cheap years does not move — moving means surrendering a rate they will not get back. So the housing stock stops turning over, the trigger for big projects never fires, and Home Depot sells caulk and paint to people staying put.

This is not a Home Depot problem. It is a rate problem wearing a retail costume, and it does not resolve on anything management can control.

The read-across

Lowe's reports into the same conditions, with consensus looking for revenue up about 9% but EPS down roughly 3% — the signature of volume held together by promotion rather than by demand.

The gap between those two lines is the more useful number in this sector right now. Revenue can be bought. Margin tells you whether the customer wanted it.

How I read it

Home Depot at these levels is a bet on the 10-year, not on retail execution. Execution is visibly fine — a 5.7% quarter in a frozen market is a good result, and management deserves the beat.

The re-rating comes when housing turnover returns, and turnover returns when mortgage rates fall enough to unlock the owners currently trapped in cheap loans. Nothing in this week's bond market suggests that is close.

What I would watch is not comparable sales. It is the ratio of large-basket to small-basket transactions, because that flips before the headline number does — and it flips the moment the housing market starts to thaw.

More market analysis by Ruslan Averin at averin.com.

NAND Revenue Rose 77% in a Quarter. Nobody Wanted the Shares

Ruslan Averin — Tue, 18 Aug 2026 23:05:05 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Combined revenue at the five largest NAND suppliers rose 77% quarter on quarter, to $68.87 billion. Samsung kept the top spot, ahead of SK Hynix and Micron.

In the same stretch, Micron shares fell 4.23% in a session and SK Hynix sank alongside them.

A 77% revenue increase in three months is not a normal number. It is what happens when a commodity is genuinely short.

The market

Metric	Value
Top-five NAND revenue, quarter	$68.87B
Growth, quarter on quarter	+77%
Leader	Samsung Electronics
Second, third	SK Hynix, Micron
Micron shares, session	−4.23%

Why good numbers get sold

Memory is not a technology story, it is a spread. Suppliers earn the gap between what a bit costs to make and what someone will pay for it, and that gap is set entirely by how tight supply is at that moment.

Which means the market does not value memory companies on trailing revenue. It values them on where the cycle is going. And a 77% quarterly jump, to most cycle-aware investors, reads as evidence that the peak is nearer — not that the trend continues.

This is the opposite of how a software business is valued, and it is why memory shares regularly fall on their best quarters and rally on their worst. Anyone applying growth-stock logic to Micron will be wrong at both ends.

What is actually short

Two demands are hitting supply at once.

Training and inference infrastructure needs high-bandwidth memory in volumes that were not planned for three years ago, and the capacity that makes it competes with the capacity that makes everything else.

Enterprise storage is absorbing the rest — the data that models are trained on has to sit somewhere, and it is not sitting on spinning disks.

Neither is a fad. Both are also fully capable of being over-served by 2028, because that is what happens every time.

The variable that decides the next eighteen months

Capital discipline among the top three.

The historical pattern is unambiguous: prices spike, all three suppliers announce expansion within a few months of each other, and the new capacity lands simultaneously about eighteen months later, killing the price. It has happened enough times that it is close to a law.

If Samsung, SK Hynix and Micron hold capacity growth below demand growth this cycle, the spread persists longer than anyone expects and the current share prices are wrong. If they do what they have always done, the shares are right and 2028 arrives with a glut.

Their capex announcements are the leading indicator, and they matter more than any quarterly revenue figure.

How I read it

I would rather own memory than accelerators here, and the reason is narrow: memory has a visible price and a visible shortage, and its shares are already down. Accelerator names are carrying both a capex-scepticism discount and a discount-rate hit while their pricing power remains an assertion.

That is a preference, not a conviction. Memory is the most brutally cyclical business in the sector, and the entry is only good while supply stays behind. The moment the top three announce expansion together, the trade is over — regardless of what revenue is doing that quarter.

More market analysis by Ruslan Averin at averin.com.

The Chips Are Falling While Memory Prices Set Records

Ruslan Averin — Tue, 18 Aug 2026 23:04:34 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

Semiconductor shares are having their worst stretch of the year, and the reason is not that anyone stopped buying chips.

SMH, the ETF tracking the 25 largest US semiconductor firms, is down 9.5% against a month earlier. In a single recent session Nvidia fell 2.89%, Broadcom 1.43%, Micron 4.23%, AMD 4.45%. The Nasdaq closed down 1.4% that day, the S&P nearly 1%, the Dow 0.7% — the sector led the market down rather than following it.

The number that does not fit

Metric	Direction
SMH, one month	−9.5%
Nvidia, session	−2.89%
Micron, session	−4.23%
AMD, session	−4.45%
NAND revenue, top five suppliers, quarter on quarter	+77%
That revenue, in dollars	$68.87B

The five largest NAND suppliers grew combined revenue 77% quarter on quarter, to $68.87 billion, while their shares were being sold. Whatever is being repriced here, it is not the top line.

What actually triggered it

TSMC beat on both profit and revenue, and simultaneously guided capital expenditure higher than previously forecast. The market sold it.

That reaction is worth understanding, because it is not irrational. Capex is a claim on future cash flow. In 2024 and 2025 the market read rising semiconductor capex as proof of demand and paid up for it. In 2026 it is reading the same disclosure as a bill that has to be paid before any of the promised earnings arrive.

The disclosure did not change. The interpretation did.

The discount rate is the other half

Semiconductor valuations are long-duration assets. The earnings that justify a forty-times multiple are not this year's — they are 2029's, and they are discounted back at a rate anchored to the long end of the Treasury curve.

That curve moved this week. The 30-year reached 5.311%, the highest since June 2007.

Run the arithmetic on any high-multiple name and a hundred basis points on the discount rate takes more off the valuation than a bad quarter would. This is why the Nasdaq fell 1.02% on 18 August while the Dow lost 0.21%: the selloff is not a sector opinion, it is a duration event that happens to hit the sector with the longest duration hardest.

The distinction that decides the trade

Two things look identical on a chart and are not:

Multiple compression with revenue intact. Demand is fine, the market is paying less per dollar of it. This resolves when rates stabilise, and the drawdown is an entry point.

The start of a demand air pocket. Orders slow, capex was built for volume that does not arrive, and the multiple compression is the market being early rather than wrong.

The NAND number argues for the first. A 77% quarterly revenue increase is not the shape of an air pocket. But NAND is memory — commodity, cyclical, currently short of supply — and it is not proof that AI accelerator demand holds up at the same time.

How I read it

I would treat memory and logic as separate trades this quarter, because the data is separating them.

Memory has visible pricing power right now and shares that are falling anyway — that is the more defensible dip. Accelerator names are being repriced on the discount rate and on capex scepticism at the same time, and only one of those two reverses quickly.

What would change my mind: a hyperscaler trimming its capital expenditure plan. That is the disclosure that turns this from a rate story into a demand story, and it would arrive in an earnings call, not in a chart.

More market analysis by Ruslan Averin at averin.com.

KKR Bid $9 Billion for a Gas Utility. The Buyer Is Really the Data Centre

Ruslan Averin — Tue, 18 Aug 2026 23:04:04 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

KKR has offered about $9 billion for UGI — $42.50 a share, a 21.1% premium to Monday's $35.09 close. Trading was briefly halted for volatility. The stock finished the session up 9.40%.

Note the gap between that and the headlines. Intraday the move was 12-13%, and several outlets ran with it. The close is what a shareholder actually owns.

The terms

Item	Value
Offer per share	$42.50
Equity value	~$9B
Premium to 17 Aug close of $35.09	21.1%
UGI close, 18 Aug	+9.40%
Status	Unsolicited, no transaction guaranteed

What is actually being bought

UGI is not a growth story. It distributes natural gas and electricity, sells propane, and owns energy infrastructure — the kind of regulated, slow, capital-heavy business that spent a decade being valued as a bond substitute and traded accordingly.

What changed is not UGI. It is the demand curve for power.

AI data centres need electricity in quantities that the grid was not built to deliver, on timelines shorter than new generation takes to permit. That has turned every existing interconnection, pipeline and distribution franchise into a scarce asset. A buyer who wants exposure to that scarcity cannot build it — the permitting alone runs years — so they buy the incumbent.

This is the same trade that has been showing up all year in different costumes: the bitcoin miner leasing 191 megawatts to an AI lab, the turbine maker with a gigawatt order book, and now a private equity firm paying a fifth over market for a Pennsylvania utility. The asset being accumulated is always power.

The 21% is smaller than it looks

A premium is measured against a market price, and the market price of utilities has been suppressed by exactly the thing making headlines this week — the 30-year Treasury at 5.311%.

Utilities compete with bonds for the same income-seeking money. When the long bond yields 5.3%, a regulated utility yielding less has to fall until it doesn't. So KKR is paying a 21% premium to a price that rising yields pushed down.

Measured against what UGI traded at when the 30-year yielded 4%, the premium is considerably less generous. Whether the board sees it that way is the whole negotiation.

What to watch

The spread. UGI closed around $38 against a $42.50 offer. That discount is the market's probability estimate. If it narrows over the coming weeks, a deal is getting more likely; if it widens, the board is resisting and the bid may go hostile or away.

A second bidder. Unsolicited approaches in scarce-asset sectors attract company. One rival approach turns 21% into a floor rather than a ceiling.

The read-across. Every listed utility with data centre load in its service territory just got repriced on this news, whether or not anyone bids for it.

How I read it

I would not chase UGI at $38 for the last four dollars of merger arbitrage — that trade pays a fixed, modest amount and carries the full downside if the board says no and the stock returns to $35.

The more interesting position is the read-across. If power distribution is worth a 21% premium to private equity, the sector is mispriced relative to what it will earn when data centre demand actually lands on those wires. That thesis does not depend on this specific deal closing.

More market analysis by Ruslan Averin at averin.com.

The 30-Year Just Printed a Yield It Has Not Seen Since 2007

Ruslan Averin — Tue, 18 Aug 2026 23:03:33 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

The 30-year Treasury reached 5.311% this week. The last time it yielded that much, the iPhone was two months old.

The 10-year sat near 4.72% in the same session. The S&P 500 closed down 0.44% on 18 August, the Dow down 0.21%, and the Nasdaq down 1.02% — a spread that tells you exactly what the bond market was doing to equity valuations.

Two forces, neither of them the Fed

Driver	Mechanism
Oil above $90	Feeds headline inflation, lifts expected policy path
Government borrowing	More supply at the long end than buyers want
Fed policy	Not the story this week

That last line is the one worth pausing on. Yields normally rise because the market expects the central bank to tighten. This move is mostly not that. It is a term premium story — the extra yield investors demand for lending far into the future when they are unsure both about inflation and about how much paper is coming.

A term premium repricing is slower and stickier than a policy repricing, because no meeting can reverse it.

What breaks at 5.3%

Three balance sheets feel this first.

Long-duration equities. A company whose cash flows arrive in 2032 is worth materially less discounted at 5.3% than at 4.3%. That is the entire mechanism behind the Nasdaq falling five times as hard as the Dow on the same day.

Anything refinancing. Debt issued during the cheap years matures into this. The coupon on the replacement is not a rounding difference — it is a permanent transfer from equity holders to lenders, and it lands on the income statement quarter after quarter.

Housing. Mortgage rates track the 10-year. Home Depot's chief financial officer described "frozen housing market conditions" this week, and the 10-year at 4.72% is a large part of why that ice is not melting.

The argument the market is actually having

Is 5.3% generous or thin?

Against 2% inflation, a 5.3% thirty-year is a good real return and the best entry point in nineteen years. Against 4% inflation — which is what oil at $91 makes plausible — it is 1.3% real for thirty years of duration risk, which is not compensation, it is hope.

Nobody knows which regime is being purchased. That uncertainty is the term premium, and it is why the yield keeps grinding higher rather than settling.

How I read it

I treat this as the primary variable and the chip selloff as a symptom. A market can absorb a bad quarter from a semiconductor company. It cannot absorb a permanent upward shift in the rate at which every future dollar is discounted.

What I would watch: whether the next new high in the 30-year comes with oil or without it. With oil, it is an energy story and reverses if the Gulf does. Without it — if yields keep rising while Brent falls — then it is the debt supply story on its own, and that one has no near-term mechanism for reversing at all.

That second scenario is the one worth being positioned for, because it is the one the equity market is not pricing.

More market analysis by Ruslan Averin at averin.com.

Selling an Apartment in Ukraine: 0%, 10% or 23%

Ruslan Averin — Tue, 18 Aug 2026 22:03:57 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

The tax on selling Ukrainian property does not depend on the profit. It depends on two facts: how many sales you have made this calendar year, and how long you owned the property.

Get those two right and the rest follows mechanically.

The three outcomes

Situation	PIT	Levy	Total
First sale of the year, owned over 3 years	0%	0%	0%
First sale of the year, owned under 3 years	5%	5%	10%
Second and later sales in the same year	18%	5%	23%

The base is the contract price, and not lower than the appraised value — a below-market price on paper does not lower the tax, it only creates a discrepancy with the valuation that the notary is required to check.

Inherited property is the notable exception: the three-year holding condition does not apply to it, so a first sale of an inherited apartment is exempt even if the inheritance was received last month.

Why the counter resets, and what it costs

The rules count sales per calendar year, not per property. An owner disposing of two apartments in the same year pays nothing on the first (if held over three years) and 23% on the second.

The same two sales split across 31 December and 2 January are two first sales, each in its own year, each potentially at 0%.

On a $100,000 apartment that scheduling decision is worth $23,000. It is the single largest tax lever available to an individual property owner in Ukraine, it requires no structure, no advisor and no risk — only the willingness to move a closing date by a few weeks.

The mechanics at the notary

The tax is settled before the contract is certified, not afterwards. The notary acts as tax agent, verifies that the payment has been made, and reports the transaction. This is why the tax question has to be answered before the deal is scheduled rather than during it.

Add the 1% pension fund levy on the purchase, customarily paid by the buyer, and the state duty. These are transaction costs rather than income taxes, but they belong in the same arithmetic when comparing a sale against holding.

Non-residents

A non-resident selling Ukrainian property is taxed at 18% plus the levy, without the exemption available on a first sale. Residency status therefore changes the outcome on the same apartment from 0% to 23% — which makes it worth establishing before the sale rather than discovering at the notary's desk.

How I read it

Ukrainian property taxation is unusual in rewarding patience twice: three years of ownership removes the income tax, and one sale per year keeps it removed.

For anyone holding more than one property, the planning is calendar work, not structuring. List the disposals you expect over the next two years, count how many fall in each, and move the ones that collide. The saving is larger than anything the sale price negotiation is likely to produce, and it is entirely within your control.

More market analysis by Ruslan Averin at averin.com.

What Actually Happens If You Do Not Declare

Ruslan Averin — Tue, 18 Aug 2026 22:03:09 +0000

Investment analysis by Ruslan Averin — originally published at averin.com.

The fine for failing to file a Ukrainian income declaration is 340 hryvnia. About eight dollars.

That number is the reason people misjudge this entirely. The declaration fine is trivial; what sits underneath it is not.

The actual structure of the exposure

Violation	Charge
Declaration not filed	340 UAH per violation
Repeated within the year	1,020 UAH
Tax understated	10% of the amount
Understatement found intentional	25%
Paid late, under 30 days	5% of the debt
Paid late, over 30 days	10%
Interest (penya)	accrues on the debt
CFC report not filed	332,800 UAH

Read that as one line rather than eight: the unpaid tax comes back with a quarter added on top, plus a late payment charge, plus interest running from the original due date — and if a foreign company is anywhere in the picture, a separate six-figure fine that has nothing to do with whether any tax was owed.

The word that costs 15 points

The gap between 10% and 25% is the finding of intent.

Intent is not established by a confession. It is inferred from conduct: income received over several years and never declared, a structure that appears designed to obscure the recipient, statements that contradict data the authority already holds. A single missed year with a plausible explanation looks different from a pattern.

This is the practical argument for fixing an omission before it is raised with you rather than after. The same unpaid tax is materially cheaper when the correction is yours.

The assumption that stopped working

The old model was that foreign income was invisible unless something drew attention to it. Automatic exchange of financial account information ended that. Participating jurisdictions transmit account balances and income of Ukrainian tax residents to Ukraine on an annual cycle, unprompted.

The tax authority does not need to find the account. It receives a file. What it does need is a reason to look at yours, and a declaration that contradicts data already in hand is exactly that.

The realistic planning assumption for 2026: the foreign account is known, the crypto-to-fiat conversions through a bank are known, and the open question is only whether your filing matches.

The time limits

The general limitation period for assessments is 1,095 days from the filing deadline. It does not run where no declaration was filed at all — an unfiled year does not quietly age out the way a filed-but-wrong one does.

At larger amounts the matter stops being purely administrative and enters criminal territory, with thresholds tied to statutory indicators and mitigation available where the assessed amount is paid. That is a lawyer's conversation, not a filing one, and the way to stay out of it is arithmetic done three years earlier.

How I read it

The cost of non-declaration is not the fine. It is the interest and the penalty accruing quietly on a liability that does not expire, in a system that now receives your account data automatically.

For a Ukrainian investor with foreign holdings, the rational position is simple and slightly boring: file every year, including the loss-making ones and the empty ones, keep the acquisition records, and treat the certificate for any foreign tax credit as a task for autumn rather than for the week before the deadline.

The tax itself, at 14% on dividends and 23% on gains, was never the expensive part. Ignoring it is.

More market analysis by Ruslan Averin at averin.com.