DEV Community: Ruslan Averin

Two Dates That Set Your Coupon: CPI on 11 September, the Fed on 16 September

Ruslan Averin — Thu, 10 Sep 2026 14:01:25 +0000

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

Two dates in the next week will decide what a US bond pays for the rest of 2026. On Friday 11 September at 8:30 ET the Bureau of Labor Statistics publishes August CPI. On Wednesday 16 September at 14:00 ET the Federal Reserve announces its rate decision, and the market currently prices roughly a 56% probability of a 25 basis point hike to 3.75–4.00%.

Everything below is built on where yields closed on 8 September and on the arithmetic of what each outcome does to them. It is not a forecast of the outcome.

The starting point

Instrument	Yield, 8 Sep 2026	Duration (approx.)
3-month bill	3.94%	0.25
2-year note	4.39%	1.9
5-year note	4.57%	4.4
10-year note	4.80%	7.9
10-year TIPS (real)	2.43%	8.5
20-year bond	5.26%	12.3
30-year bond	5.25%	15.0
IG corporate	5.53%	6.8
BB high yield	6.11%	3.8

July CPI was 3.4%. Every Treasury maturity from one month out yields more than that in nominal terms, and every maturity from one year out yields more than a point above it. That has not been true for most of the past fifteen years.

Scenario one: the Fed hikes

A 25 basis point hike on 16 September lifts the front end almost one-for-one. The three-month bill moves to about 4.2%, the two-year — which already prices most of one hike — perhaps 10–15 basis points higher. The long end is the interesting part. A hike that the market reads as credible — a Fed willing to act on 3.4% inflation — flattens the curve: the 30-year may not rise at all and could fall, because the inflation premium that pushed it to 5.25% shrinks. A hike read as too little does the opposite and steepens the curve further.

Positioning for scenario one: short duration is safe and long duration is a bet on Fed credibility. The two-year at 4.39% loses about 0.3% of price on a hike and earns that back in a month of coupon.

Scenario two: the Fed holds

A hold with hawkish language — the "we are watching September CPI" outcome — is the scenario the long end fears most. It keeps the policy rate 1.5% below the 30-year yield, keeps the inflation question open, and gives the market another six weeks to sell duration into every auction. In that scenario the 10-year retests 4.85% and the 30-year retests 5.31%.

A hold with soft language — a Fed signalling that the July and August payroll numbers of 21,000 and 162,000 worry it more than the CPI — would rally the front end and do very little for the back.

What a hot CPI print does to both

The Friday CPI comes first, and it can change the odds for Wednesday. Consensus is for a headline print close to July's 3.4%. A print at 3.6% or above pushes hike odds toward 75% and adds 10–15 basis points to the two-year on the day. A print at 3.2% or below pulls the odds under 40% and the two-year back toward 4.25%. The 30-year moves less on either, because the long end is trading supply and credibility, not the monthly print.

The ladder that works in both scenarios

For a private investor who is not trading the outcome but wants the yield, the structure that survives both scenarios is a short-to-intermediate ladder with an inflation-protected anchor:

Bucket	Instrument	Share	Why
0–1 year	Bills, 3.94–4.15%	30%	Reprices up on a hike, no principal risk
2–3 years	Notes, 4.39–4.44%	30%	Locks most of the curve's rise, duration under 3
5–10 years real	TIPS, 2.17–2.43% real	25%	Removes the inflation bet that drives the long end
10–20 years	10Y 4.80% / 20Y 5.26%	15%	Long coupon, sized so a 0.5% move costs under 1% of the total

What it excludes: the 30-year (the 20-year pays the same with less duration), investment-grade corporates (the 0.81% spread does not pay for a widening), and anything below BB.

For a non-US investor: the tax line that changes the answer

This is the part that a Ukrainian or European reader should not skip, because it moves the net yield by more than any Fed decision.

Interest on US Treasuries paid to a non-resident is exempt from US withholding tax under the portfolio-interest rules, provided the broker has a valid W-8BEN on file. A Ukrainian resident who buys a Treasury note directly through a broker receives the full 4.39% coupon with nothing withheld in the US. The same applies to most registered corporate bonds.

Distributions from a US-domiciled bond ETF are dividends, not interest. They are subject to 30% withholding, reduced to 15% under a tax treaty — Ukraine's treaty with the US provides 15%. Some US ETFs designate part of their distributions as "qualified interest income" and pass Treasury interest through without withholding, but not every fund does and not every broker applies it. A 4.39% yield becomes 3.73% after a 15% haircut and 3.07% after 30%.

An Irish-domiciled UCITS ETF holding US Treasuries pays no US withholding on the interest at fund level and is the usual route for European investors who want the yield in an ETF wrapper. Accumulating share classes defer the income entirely.

What none of this removes is the home-country tax. A Ukrainian resident's foreign investment income — coupons and capital gains — is taxed at 18% personal income tax plus the 5% military levy, and must be declared. That 23% is the real cost, and it applies equally to a direct Treasury and to an ETF. It is also the reason hryvnia OVDP, which carry no income tax and no levy, remain competitive on a net basis despite the currency risk.

The one-line version

Own the front end for the coupon, own TIPS for the real return, keep the long end small enough that Friday and Wednesday cannot hurt you, and buy Treasuries directly rather than through a US ETF if you live outside the United States. The curve overview has the full set of yields this rests on.

This is analysis, not tax advice. Withholding rates depend on residence, treaty status and broker documentation; confirm them for your own situation.

More market analysis by Ruslan Averin at averin.com.

Corporate Bonds Pay 5.53%. Only 0.81% of That Is for Corporate Risk

Ruslan Averin — Thu, 10 Sep 2026 14:00:54 +0000

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

An investment-grade corporate bond index yielded 5.53% on 8 September 2026. That is a good number in absolute terms — a year ago it was 4.79%. It is a bad number once it is taken apart.

Of the 5.53%, 4.72% is the yield on the Treasury bond of matching maturity. The remaining 0.81% is the option-adjusted spread — the compensation an investor receives for lending to a corporation rather than the US government. Eighty-one basis points a year is what the market charges for the possibility of default, downgrade, illiquidity and every other thing that can go wrong with a company and cannot go wrong with the Treasury.

The corporate bond's yield went up 0.74% in twelve months. The Treasury's went up 0.70%. The spread barely moved. Corporate bonds are not paying more for corporate risk this year; they are paying more because government bonds are.

The spread ladder, rung by rung

Segment	Effective yield	Spread over Treasuries	Treasury component
AAA corporate	5.40%	0.43%	4.97%
Investment grade (all)	5.53%	0.81%	4.72%
BBB	5.71%	0.99%	4.72%
BB	6.11%	1.55%	4.56%
High yield (all)	7.22%	2.67%	4.55%
B	7.31%	2.76%	4.55%
CCC and below	15.01%	10.56%	4.45%

Source: ICE BofA indices via FRED, 8 September 2026.

Two things stand out. First, the investment-grade spread of 0.81% is within a few basis points of its two-year low (0.75% in June 2026), and the high-yield spread of 2.67% is likewise near the bottom of its range. These are not distressed levels; they are the levels of a market that is not worried. Second, the ladder is not linear. From BB to B the spread widens by 1.21%. From B to CCC it widens by 7.8%. The market is pricing the bottom of the credit stack as a different asset class, and it is right to.

What 0.81% has to cover

The long-run average annual default rate for investment-grade credit is a few tenths of a percent, and recovery on a defaulted senior bond has historically been around 40 cents on the dollar. Expected loss on a diversified IG portfolio is therefore small — call it 0.1–0.2% a year in a normal year. On that arithmetic 0.81% looks like adequate compensation.

The problem is that the spread is not only a default premium. It also has to pay for:

Downgrade risk. A BBB bond that becomes BB is sold by every fund with an investment-grade mandate, and the holder eats the price gap between 0.99% and 1.55% of spread — roughly 3% on a seven-year bond — without any default occurring.
Liquidity. Corporate bonds do not trade like Treasuries. In March 2020 the IG spread went from 1% to 4% in three weeks, and the bid disappeared for anything that was not a benchmark issue.
Supply. Corporate issuance in 2026 is running above $1.5 trillion, driven by AI capital expenditure. The hyperscalers are excellent credits, but the sheer volume of paper is what keeps spreads from tightening further and is what will widen them if the AI capex cycle turns.

Eighty-one basis points covers the first item comfortably and the second and third barely at all. That is a description of a spread priced for a good year, not a bad one.

High yield: 7.22% for a reason

High yield is the more honest market. A 7.22% yield with a 2.67% spread says that the index expects some defaults and is being paid for them. Within it, BB at 6.11% is the sweet spot for an investor who wants more than Treasuries without taking company-specific risk: the historical default rate for BB is low, the spread of 1.55% is nearly double investment grade, and the bonds are mostly larger issuers one notch below the line.

CCC at 15.01% is not a yield. It is a probability. A bond that yields 15% when the risk-free rate is 4.45% is pricing a substantial chance that the coupon will not be paid in full, and the 10.56% spread is the market's estimate of the expected loss plus the premium for being the one to absorb it. Some CCC credits will pay in full and return 15%; the index will not.

What I would own

The decision this month is not IG versus HY. It is Treasuries versus credit, and Treasuries win most of the argument.

Investment grade at 0.81%: own it only where a specific mandate requires corporate paper. The marginal 0.81% over Treasuries does not pay for a spread-widening event, and there will be one.
BB at 6.11%: the one segment of credit where the extra 1.55% is reasonable compensation for the incremental risk, provided the position is diversified across issuers.
B and CCC: trading instruments for specialists. Not an allocation.
The Treasury alternative: the 10-year at 4.80% and the 20-year at 5.26% offer most of the corporate yield with none of the corporate risk, and TIPS at a 2.43% real yield remove the inflation risk that is the actual reason yields are where they are.

The corporate bond market in September 2026 is paying investors well for government risk and poorly for corporate risk. Take the yield where it is being offered. The curve overview has every Treasury maturity; the playbook turns this into a portfolio.

More market analysis by Ruslan Averin at averin.com.

The 30-Year Treasury at 5.2%: Highest Since 2001, and Who Is Still Buying

Ruslan Averin — Thu, 10 Sep 2026 14:00:23 +0000

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

On 14 August 2026 the US Treasury sold $25 billion of 30-year bonds at 5.216%. It was the highest yield at a 30-year auction since 2001. Three days later the secondary market printed 5.31%, the highest daily close in the FRED series since 2007. On 8 September the bond closed at 5.25%.

A quarter of a century is a long time. The last time the US government paid this much to borrow for thirty years, the policy rate was 6.5%, the federal debt was $5.7 trillion, and the deficit had just turned into a surplus. Today the policy rate is 3.75%, the debt has crossed $40 trillion, and the 12-month deficit is $1.8 trillion. The 30-year yield is not high because short rates are high. It is high in spite of them.

Who bought the auction

The auction statistics are the best window into who is still willing to lend for thirty years.

30-year auction, 14 August 2026
Size	$25 billion
High yield	5.216%
Bid-to-cover	2.39
Primary dealers' share	11.5%

A bid-to-cover of 2.39 is on the low side of the past two years' range. The dealer share of 11.5% is the number to watch: primary dealers are obliged to bid, and the paper they absorb is the paper nobody else wanted. At 11.5% they were not left holding the bag, but the buyer base is narrower than it was. Domestic funds and insurers took the rest.

Compare the 10-year auction on 9 September: 4.834%, bid-to-cover 2.71, with 79.2% going to indirect bidders — the category that includes foreign central banks and overseas funds. Demand for ten-year paper is fine. Demand for thirty-year paper is the problem, because the marginal foreign buyer has stopped showing up at the long end.

Who is selling

Two sellers matter, and one of them is a government.

Japan's Ministry of Finance has been selling Treasuries repeatedly through the summer to defend the yen. When Tokyo needs dollars to buy yen, it sells the most liquid dollar asset it holds, and it holds a great deal of long US paper. Every intervention is a bid removed from the 30-year.

The second seller is the US Treasury itself, and it is selling by necessity. Debt held by the public reached 100% of GDP in August. Net interest is running at roughly $1.25 trillion a year, more than defence. The deficit is $1.8 trillion over twelve months and $168 billion in August alone. That has to be financed, and Treasury has chosen to finance a growing share of it in coupons rather than bills. The one mitigating move — tripling the buyback programme to $6 billion of off-the-run long bonds — is a rounding error against $25 billion auctions every month.

Behind both sits the private-sector flood: corporate issuers, led by hyperscalers borrowing to fund AI capex, have sold more than $1.5 trillion of bonds this year. A pension fund choosing between a Microsoft 30-year and a Treasury 30-year does not have to buy the Treasury.

The arithmetic of a 30-year bond at 5.25%

This is where the coupon stops being the point. A 30-year bond with a 5.25% coupon at par has a modified duration of about 15. That means:

Yield move	Approximate price change
+0.25%	−3.7%
+0.50%	−7.1%
+1.00%	−13.5%
−0.50%	+8.0%
−1.00%	+16.9%

A half-point rise in yield — the distance the 30-year travelled between May and August — costs a holder seven percent of principal, or sixteen months of coupon. The bond pays 5.25% a year and can lose that in a fortnight.

Set that against the 2-year at 4.39% with a duration under two: the same half-point move costs under 1%. The 30-year offers an extra 0.86% of yield for roughly eight times the price risk. In September 2026 that trade is being paid, but not generously.

Where the value is on the long end

If the objective is the highest coupon per unit of duration, the answer is not the 30-year. The 20-year at 5.26% pays a basis point more with a duration of about 12.3 instead of 15. It trades cheap because it has fewer index buyers, which is precisely why a buyer who does not care about the index should own it.

If the objective is a locked real return for a long horizon, the 10-year TIPS at a 2.43% real yield is the cleaner instrument. It carries the same government credit, removes the inflation risk that is the main reason the long end has repriced, and for a buyer who will hold to maturity is very close to what a 30-year at 5.25% is supposed to deliver — without the 15-duration exposure to the next Japanese intervention or the next hawkish Warsh speech.

Is 5.25% a top?

Nobody knows, and the honest range is wide. The case for a top: 5.25% nominal is 1.85% above July CPI and 1.5% above the policy rate, the auction cleared, and a Fed hike on 16 September would flatten the curve by pulling the front end up faster than the back. The case against: the supply is structural, the foreign bid is shrinking, the deficit is not on any path to $1 trillion, and a 3.4% inflation rate with oil elevated is not a rate at which a 30-year bond is obviously cheap.

My own position is that the 30-year is a trading instrument in 2026, not a holding. The yield is attractive enough to buy on a move above 5.3% and sell on a move below 5%. It is not attractive enough to lock away for three decades when the 20-year pays the same coupon and TIPS lock the real return without the inflation bet. The yield-curve overview shows where the 30-year sits against the rest of the curve; the playbook turns this into an allocation.

More market analysis by Ruslan Averin at averin.com.

The Fed Held at 3.75%. The Bond Market Hiked Anyway: The US Yield Curve in September 2026

Ruslan Averin — Thu, 10 Sep 2026 13:59:53 +0000

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

The Federal Reserve has not raised interest rates in 2026. It cut three times in the autumn of 2025, parked the target range at 3.50–3.75% in December, and has held it there at every meeting since. Over the same nine months the 10-year Treasury yield went from 4.19% to 4.80%, and the 30-year from 4.86% to 5.25%.

That is the whole story of the US bond market this year in two sentences. The central bank stood still and the market tightened for it. Anyone who bought duration in January on the logic that "the Fed is done, yields fall next" has lost money on the price and is being paid a coupon that no longer covers the loss.

What every maturity pays on 8 September 2026

Maturity	5 Sep 2025	8 Sep 2026	Change
1 month	4.29%	3.81%	−0.48%
3 months	4.07%	3.94%	−0.13%
6 months	3.85%	4.00%	+0.15%
1 year	3.65%	4.15%	+0.50%
2 years	3.51%	4.39%	+0.88%
3 years	3.48%	4.44%	+0.96%
5 years	3.59%	4.57%	+0.98%
7 years	3.80%	4.68%	+0.88%
10 years	4.10%	4.80%	+0.70%
20 years	4.72%	5.26%	+0.54%
30 years	4.78%	5.25%	+0.47%

Read the table top to bottom and the shape of the year is visible. The one-month bill is the only maturity that pays less than a year ago, because it tracks the policy rate and the policy rate was cut. Everything from one year out pays more, and the two-to-five-year sector pays almost a full percentage point more. A year ago the curve was inverted at the front: bills at 4.29% and the two-year at 3.51%. Today it is upward-sloping from end to end.

Why the curve steepened without the Fed

Three forces pushed yields up while the policy rate sat still, and none of them is about to reverse.

Inflation stopped falling. Headline CPI was 3.4% in July 2026, with the energy index up 14.7% year over year on oil prices that have not come down since the Iran war began. Core CPI, at 2.5%, looks better, but the Fed does not get to eat only the core. The market has watched the "last mile" of disinflation stall for a year and has repriced accordingly.

The Fed's own chairman told the market to expect a hike. Kevin Warsh's Jackson Hole address in late August was read as a signal that the next move is up, not down. Rate futures moved from pricing cuts to pricing roughly a 56% chance of a 25 basis point hike at the 15–16 September meeting. Two-year yields, which are essentially a bet on the next two years of policy, rose from 3.47% at the start of the year to 4.39%.

Supply. Debt held by the public crossed $40 trillion and reached 100% of GDP in August. The 12-month deficit is running at $1.8 trillion. Treasury has to sell that, and it is selling it into a market where Japan has been a net seller to defend the yen and where corporate issuers — AI capex borrowers above all — have brought north of $1.5 trillion in new paper this year. Every one of those buyers competes for the same pool of savings, and the long end pays the price.

The real yield is the number that matters

Strip inflation out and the picture is unusual for the post-2008 world. The 10-year TIPS real yield is 2.43%. The 10-year breakeven inflation rate — the gap between the nominal and the inflation-protected bond — is 2.37%. Together they add to the 4.80% nominal.

For most of the 2010s the 10-year real yield was below 1%, and for stretches it was negative. A 2.4% real yield means a buyer who holds to maturity is promised 2.4% a year above whatever inflation turns out to be. That is a genuine return, and it is the reason the bond market has become interesting again rather than simply painful.

The catch is the same as it always is: the real yield is only locked for the holder who does not sell. Anyone marking to market is exposed to the next move in nominals, and the next move has been up for nine months.

The kink at the long end

One detail in the table deserves its own line. The 20-year Treasury yields 5.26% and the 30-year 5.25%. The longest bond on the curve pays a basis point less than the one ten years shorter. That is not a forecast; it is a supply artefact. The 20-year was reintroduced in 2020, has fewer natural buyers than the 30-year, and trades cheap to the curve most of the time. But it means an investor who wants the maximum coupon per unit of duration should look at the 20-year, not the 30-year — a point I return to in the 30-year piece.

What the shape says

A steep, upward-sloping curve with a 10-year/2-year spread of +0.40% and a 10-year/3-month spread of +0.88% is the textbook late-cycle steepener: short rates anchored by a central bank that is not sure which way to move, long rates lifted by inflation risk and supply. It is not a recession signal. The inverted curve of 2023–2025 was the recession signal, and the recession did not arrive.

What the shape does say is that the market no longer believes the Fed can cut its way out of a 3.4% inflation print with oil where it is. The cuts of 2025 are, in the bond market's judgement, finished. Whether the next move is a hike on 16 September or a hold with hawkish language, the long end has already priced a Fed that is done easing.

Three other pieces this week take the same data further: the 30-year at 5.2% and who is still buying it, corporate credit and whether 0.81% is enough for taking company risk, and the practical playbook for the two dates that will set the coupon for the rest of the year.

More market analysis by Ruslan Averin at averin.com.

Ruslan Averin: Corporate Bonds Outperform European Rental Real Estate

Ruslan Averin — Sun, 06 Sep 2026 23:12:57 +0000

In an evolving macroeconomic framework shaped by central bank rate stabilization, private investor and financial analyst Ruslan Averin has published a comparative market analysis evaluating liquidity profiles and risk-adjusted yield spreads between European prime residential real estate and investment-grade (IG) corporate credit markets.

Analyzing portfolio performance metrics across Western Europe, Averin highlights a growing structural yield inversion. Net capitalization rates for residential buy-to-let portfolios in primary metropolitan hubs—including Frankfurt, Paris, and Amsterdam—have compressed to an average net yield of 2.8% to 3.9%, once accounting for local property transfer taxes (ranging from 6% to 10.4%), ongoing municipal levies, maintenance reserves, and property management overheads.

Conversely, investment-grade corporate bond funds and euro-denominated fixed-income instruments currently provide benchmark coupon yields between 4.8% and 6.2% with zero property-level operational friction.

"Institutional capital allocation decisions are fundamentally dictated by the risk-adjusted spread over risk-free rate benchmarks like 10-year German Bunds," states Ruslan Averin. "When prime European residential real estate offers a narrow spread of barely 100 to 150 basis points over sovereign debt—while carrying severe illiquidity and regulatory risk—it fails the capital efficiency test. Investment-grade corporate credit yields deliver a far more attractive cash-flow distribution."

In his commentary, Averin outlines three core structural advantages of corporate fixed-income allocation over physical real estate portfolios:

Asymmetric Capital Efficiency: Corporate credit delivers clean cash yields without cap-ex drag, municipal property taxes, or landlord compliance costs associated with European energy efficiency mandates (EPC ratings).
Institutional Market Liquidity: Real estate assets require 90 to 180 days on average to execute a settlement, incurring broker and legal fees of up to 5%–8%. In contrast, corporate bond portfolios can be rebalanced or liquidated on secondary markets instantly.
Mitigation of Regulatory Tail-Risk: Residential property markets in key European jurisdictions face increasing political pressure through rent freezes, eviction moratoriums, and strict statutory caps, whereas corporate debt instruments operate under transparent balance-sheet covenants and clear maturity schedules.

"Flexibility and capital velocity are paramount components of modern risk management," Averin concludes. "Allocating capital into illiquid, highly regulated physical real estate for low single-digit net returns is sub-optimal when high-grade corporate credit offers a superior risk-adjusted yield curve with instantaneous liquidity."

For further research reports, macroeconomic analysis, and portfolio commentaries, visit the official resource at https://averin.com

Rent or Coupons? Part 3 (Final): The Verdict

Ruslan Averin — Sun, 06 Sep 2026 23:10:22 +0000

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

Two parts, two defensible answers. Part 1 showed a Kyiv one-bedroom netting 4.35% against 5.51% for investment grade corporate credit. Part 2 showed a credit market paying 0.81% for corporate risk, a subsidised mortgage at 3–7%, and a tax-free exit after three years.

Both are true. They are answers to different questions, and the whole dispute is caused by asking the wrong one — which asset is better instead of what is this capital for.

The numbers on one page

	Kyiv one-bedroom, $68,500	Investment grade USD credit	Hryvnia OVDP
Gross yield	7.36%	5.51%	15.18–16.47%
Net of tax and costs	4.35%	~5.5% before residence tax	15.18–16.47%, no tax
Currency of income	Dollar-linked	Dollar	Hryvnia
Liquidity	Months	One day	One day
Leverage available	3–7% via єОселя	None	None
Can the issuer rewrite terms?	No	Only in default	Yes — 2024 precedent
Work required	Continuous	None	None
Exit tax	0% after 3 years, first sale	Taxed	Taxed on gain

The break-even that settles it

The apartment nets 4.35%. Investment grade pays 5.51%. The gap is 1.16% a year, and that is precisely the appreciation the apartment must deliver, every year, simply to draw level.

Over the last twelve months it delivered about 5% — it beat the bond, and comfortably. But the rent that funds the coupon side of the comparison fell 12% in dollars over the same period. An asset whose price rises while its income falls is being bought for capital gain, not for yield, and it should be assessed as a capital gain bet, honestly and out loud.

So the real question is not 4.35% against 5.51%. It is: does the buyer believe Kyiv residential prices rise more than 1.16% a year, on average, over the holding period? Anyone who does should buy the apartment. Anyone who does not is buying a bond with a plumbing obligation attached.

Three questions that decide it, in order

1. Horizon. Entry costs on a Kyiv apartment run to roughly 4% and exit costs to another 3–5% between the agent and the negotiation. Under five years, that 7–9% round trip eats the entire yield advantage before the argument begins. Under five years the answer is fixed income; the property comparison does not even need to be run.

2. Leverage. Whether the buyer qualifies for єОselya at 3% or 7% changes the answer more than any yield in this series. With subsidised financing, the tenant amortises someone else's principal and inflation erodes the debt — that is a materially different asset from the same apartment bought for cash. Without it, the apartment is an unlevered 4.35%, and the bond wins on the arithmetic.

3. Currency of liabilities. Someone who spends hryvnia and buys a dollar coupon is running a currency mismatch and calling it prudence. Someone who spends dollars and holds hryvnia OVDP at 16.47% is being paid handsomely to run the same risk in reverse. Match income to spending first, then optimise the yield.

Where the government bond and the corporate bond diverge

The series has treated bonds as one asset. They are two.

Government paper — Ukrainian OVDP at 15.18–16.47% tax-free, or US Treasuries at 4.34–5.25% — is a position on the state's willingness and ability to pay. In Ukraine that comes with 122% debt-to-GDP and a 2024 restructuring in living memory; in the US it comes with duration and an FOMC meeting on 15–16 September where the market prices roughly a 58% chance of a hike.

Corporate paper — investment grade at 5.51%, high yield at 7.15%, Ukrainian hryvnia issuers up to 18% — pays a spread of 0.81% and 2.65% respectively for taking default risk on top of that. Those spreads are historically thin. Investment grade at 81 basis points is compensation for credit risk that barely exists in the price, which is a reason to own the government bond and skip the corporate one, not a reason to reach for high yield.

The single cleanest instrument in this entire comparison, for a Ukrainian resident, remains hryvnia OVDP: the highest nominal yield on the page, no income tax, no military levy, daily liquidity — carrying, in exchange, exactly one risk that the apartment does not have, and it is the issuer.

The allocation this argument actually supports

The framing that survives all three parts is not a choice between two assets, it is an ordering:

Liquidity and short horizons — under five years, any currency: bonds, and preferably government ones. There is no version of this comparison where an apartment wins over three years.
Hryvnia income and hryvnia spending — OVDP first, before deposits and before corporate issuers. Tax-free 15–16.5% is the best risk-adjusted line available domestically, and the risk is the sovereign, which the buyer already carries by living there.
Dollar income, long horizon, no leverage available — investment grade credit or Treasuries; take the government yield and decline to be paid 0.81% for corporate risk.
Access to єОselya at 3–7%, a horizon over seven years, and a tolerance for the work — the apartment, bought below the district median, renovated, and held past the three-year mark for the 0% exit.
Cash purchase at the median price, held passively, no renovation, no leverage — the worst version of both ideas. It carries property risk, property illiquidity and property work in exchange for a yield lower than a Treasury bill's.

The verdict

The landlord and the bondholder rarely disagree about facts. They disagree because the landlord quotes 7.36% and the bondholder quotes 5.51%, and only one of those numbers is net.

The honest comparison is 4.35% against 5.51%, and it says: without leverage, without a renovation plan and without a seven-year horizon, the coupon wins. With any two of the three, the apartment wins, and it wins on the parts of the return that a bond structurally cannot offer — a resetting income, an asset that cannot be restructured, and an exit that goes untaxed.

What no version of the arithmetic supports is the default behaviour: buying at the median for cash because property feels safer than paper. In September 2026 it yields less, moves slower, and demands more.

More market analysis by Ruslan Averin at averin.com.

Rent or Coupons? Part 2: The Case Against the Bond

Ruslan Averin — Sun, 06 Sep 2026 23:09:51 +0000

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

Part 1 valued a Kyiv apartment the way an analyst values a bond: rent as coupon, price as par, and a net yield of 4.35% against 5.51% for investment grade corporate credit. On that arithmetic the apartment loses.

The arithmetic is right. The framing is the problem — because it silently treats a bond as risk-free and an apartment as an inefficient bond, and in September 2026 both halves of that assumption are wrong.

What the credit market is actually paying for risk

The yields in Part 1 look generous. The spreads behind them do not.

Measure	Level, 3 September 2026
Investment grade corporate index yield	5.51%
Investment grade option-adjusted spread	0.81%
High yield index yield	7.15%
High yield option-adjusted spread	2.65%
10-year Treasury	4.77%
10-year inflation-protected real yield	2.42%

An investment grade spread of 81 basis points means the market is charging less than one percent a year to take corporate default risk over Treasuries — a level not sustained since before the financial crisis. High yield at 2.65% is being paid to lend to companies that, by definition, may not repay.

Nearly all of the 5.51% is not credit compensation at all. It is the government yield, and it comes with the government yield's exposure: duration. On an index with roughly seven years of duration, a one percentage point rise in yields costs about 7% of principal — two years of coupon, gone on a repricing, which is exactly what the long end delivered this summer when the 30-year touched 5.31%, its highest since 2007.

The FOMC meets on 15–16 September with markets pricing roughly a 58% probability of a 25 basis point hike, not a cut. A bondholder is being paid 0.81% to carry corporate risk into that meeting.

An apartment is a real asset; a bond is a nominal promise

The coupon is fixed in the currency it was written in. Rent is not: it resets with every lease.

That distinction is invisible in a one-year comparison and dominant over ten. The 10-year TIPS real yield is 2.42% — that is the honest, inflation-adjusted return on a Treasury today, before tax. A hryvnia OVDP paying 15.5% against an NBU inflation forecast near 10% is a real 5%, and it is a real 5% only for as long as inflation behaves and the hryvnia holds.

The apartment carries no such promise, and needs none: it is priced in square metres, and square metres are re-priced by the market every year. Over the last twelve months that mechanism did exactly what it is supposed to do — the median Kyiv one-bedroom rose about 5% in dollar terms while rents fell. The asset absorbed the shock; the income line took it.

Sovereign risk cuts both ways

The comparison in Part 1 credited Ukrainian government bonds with a tax-free 15.5%. It did not price what stands behind them: debt at roughly 122% of GDP, a budget financed by external partners, and a country at war. Ukraine restructured its external commercial debt in 2024, and bondholders took a substantial writedown by agreement of a creditor committee.

That is the difference in kind between the two assets. A sovereign obligation can be rewritten by the issuer. An apartment cannot be restructured by anyone. A corporate bond in default recovers something in the region of 40 cents; the building does not go to zero, because someone still needs to live in it.

The leverage nobody offers a bondholder

The strongest argument for property is not the yield. It is the financing.

Ukraine's єОселя programme lends at 3% to serving military, medics and teachers, and at 7% to everyone else within its criteria. Against a gross rental yield of 7.36%, a 7% mortgage is roughly break-even on carry — and roughly break-even carry on an appreciating asset bought with someone else's money, amortised by a tenant, in a currency losing 2–4.5% of value a year, is a very different proposition from an unlevered 4.35%.

At the subsidised 3% rate it is not close. No broker will lend a retail investor at 3% to buy corporate bonds.

The coupon cannot be renovated

This is the asymmetry no fixed income instrument has: the owner can act on the asset.

A $6,000–8,000 renovation moves a Kyiv one-bedroom out of the median rent band. Turning $420 a month into $550 adds roughly $1,560 a year on $7,000 spent — an incremental return north of 20% on the incremental capital, and it lifts the resale price at the same time.

The same logic applies at purchase. The median is not the market: one-bedrooms run from $43,500 in Desnianskyi to $152,000 on Pechersk. Buying 15% below the district median, which is ordinary competence in a slow market, changes the entry yield from 7.36% to about 8.7% before anything else is done. Nobody buys a corporate bond 15% below its market price.

Taxes favour the building on exit

Part 1 counted the tax on rent and stopped. The exit matters more.

A first sale of residential property in a calendar year, held for over three years, is taxed at 0% in Ukraine — no income tax, no military levy. Ukrainian corporate bond coupons are taxed at 18% plus the 5% levy. So the two assets are taxed in mirror image: the bond is taxed on its income and the apartment is taxed on its income, but the apartment's capital gain — the part that compounds — can be realised tax-free, and the bond has no capital gain to realise.

How I read it

Part 1 is honest about the income and blind to everything else. The apartment yields less, and in exchange the owner gets an asset whose value resets with inflation, that cannot be restructured by an issuer, that a state programme will finance at 3–7%, that responds to work, and that can be sold tax-free after three years.

The bond gives 5.51% for taking corporate risk at 0.81% of compensation, into a meeting where the market thinks the Fed hikes.

Neither of these is the answer. Part 3 puts both cases against each other and works out where the line actually falls — by horizon, by leverage, and by currency.

More market analysis by Ruslan Averin at averin.com.

Rent or Coupons? Part 1: Why the Kyiv Landlord Is Losing to a Bond

Ruslan Averin — Sun, 06 Sep 2026 23:09:20 +0000

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

Every argument about buying an apartment to rent out starts with the same number, and it is the wrong one.

In Kyiv in September 2026 the median one-bedroom on the secondary market sells for $68,500 and the median one-bedroom rents for UAH 19,000 a month — about $420. Twelve months of that rent against the purchase price gives a gross yield of 7.36%. Set beside a deposit or a coupon, 7.36% looks decisive, and most buying decisions stop there.

It should not. Gross rent is a revenue line, not a return. The comparison that matters puts the net rental yield against the net coupon, and once the deductions are applied the ranking changes.

What survives the deductions

The costs below are the ordinary ones a Kyiv landlord meets in a normal year, not a disaster scenario. The tax figures assume the common route — an individual entrepreneur on the third group, which is the cheapest legal way to declare rent in Ukraine. A private individual declaring rent outside that regime pays 18% income tax plus the 5% military levy, and the arithmetic gets worse, not better.

Item	Annual
Gross rent, 12 months at $420	$5,040
One month vacancy between tenants	−$420
Single tax, 5% of receipts	−$231
Military levy, 1% of receipts	−$46
Unified social contribution, UAH 1,902.34 a month	−$512
Maintenance and depreciation, 1% of value	−$685
Agency fee on tenant turnover, amortised	−$105
Utilities and building fees while vacant	−$60
Net income	$2,981
Net yield on $68,500	4.35%

Nothing in that table is aggressive. One vacant month a year is standard. One percent of the property value for maintenance is the number that lets an apartment be re-let at the same rent in year five as in year one; spend less and the rent drifts down instead.

And the table ignores the entry cost entirely: the 1% pension fund levy on purchase, the notary, the agent. Together they run to roughly 4% of the price — paid once, but paid before the first coupon-equivalent ever arrives.

What the same capital earns as a coupon

The same $68,500, in fixed income, on 3 September 2026:

Instrument	Yield
US 2-year Treasury	4.34%
US 10-year Treasury	4.77%
US 30-year Treasury	5.25%
ICE BofA US investment grade corporate index	5.51%
ICE BofA US high yield index	7.15%
Ukrainian USD-denominated OVDP	about 4.2%
Ukrainian hryvnia OVDP, 1 year (1 September auction)	15.18%
Ukrainian hryvnia OVDP, 3 year (25 August auction)	16.47%
Ukrainian corporate bonds, hryvnia	up to 18%

The ten-year obligation of the United States government pays more than a Kyiv apartment nets its owner. Investment grade corporate credit pays 5.51% — 1.16% more than the apartment, in dollars, settling in a day, requiring no tenant, no plumber and no notary. High yield pays 7.15%: more than the apartment's gross yield, before a single deduction.

That is the whole case, and it is arithmetic rather than opinion.

The hryvnia instrument nobody prices properly

The most mispriced line in that table is the Ukrainian government bond. Hryvnia OVDP placed between 15.18% and 16.47%, and income from them carries no personal income tax and no military levy — the only instrument available to a Ukrainian retail investor where the state takes nothing.

Rent is taxed. Corporate bond coupons are taxed. Deposits are taxed. OVDP are not.

Against the NBU forecast of about 10% inflation for the end of 2026, a 15.5% coupon is a real return of roughly 5%. Against the hryvnia — which held near UAH 44.6 to the dollar through the year, with forecasts of UAH 45.5–46.7 by December — a full year of expected devaluation costs somewhere between 2% and 4.5%. Even at the pessimistic end, the dollar-equivalent return lands around 11%, against 4.35% net from the apartment.

The direction of travel

The static comparison is unflattering. The trend is worse.

Over the past year the median Kyiv one-bedroom rose about 5% in price, while the median one-bedroom rent fell 5% in hryvnia and 12% in dollars. Numerator down, denominator up: that is yield compression, and it is mechanical. Every month that combination persists, the buyer's entry yield gets thinner.

By district the fall is sharper than the median suggests — one-bedroom rents dropped 17% in Dniprovskyi and 11% in Shevchenkivskyi and Darnytskyi. Meanwhile the price spread across the city runs from $43,500 in Desnianskyi to $152,000 on Pechersk, a 3.5-fold range on the same nominal product.

What a coupon does not ask for

The yield gap is only part of it. The apartment also charges its owner in things that never appear in a yield calculation:

Time. Tenants, viewings, repairs, disputes, tax filings. A bond has never once called about a leaking pipe.
Concentration. $68,500 in one object, one district, one city, one jurisdiction, one currency of law. The same money buys twenty corporate credits across ten sectors.
Liquidity. A listed bond settles in a day. A Kyiv apartment takes months to sell, and the last 5% of the price is normally left on the negotiating table.
Optionality. Cash in liquid instruments can be redeployed when something breaks. Cash in an apartment cannot.

How I read it

The case for the coupon in September 2026 does not rest on the apartment being a bad asset. It rests on a comparison that landlords rarely run honestly: 7.36% gross against 5.51% net is not a comparison at all. The correct one is 4.35% against 5.51%, and on that line the apartment loses — while asking for work that the bond does not.

Which is not the end of the argument. Everything above prices the apartment as a bond, and an apartment is not a bond. Part 2 makes the opposite case, and it is stronger than this one looks.

More market analysis by Ruslan Averin at averin.com.

Asia Borrowed $16 Billion for AI in One Week

Ruslan Averin — Fri, 28 Aug 2026 22:46:09 +0000

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

In one week, two Asian technology companies went to the market for roughly $16.5 billion, both explicitly to fund artificial intelligence. Both stocks fell on the news.

Alibaba placed 710 million new shares at HK$112.70 for HK$80 billion — about $10.2 billion, and the largest primary follow-on ever by a Hong Kong-listed company. SoftBank filed to issue ¥1 trillion of seven-year retail bonds, roughly $6.3 billion, at an indicative coupon of 4.3% to 4.9%.

Two ways to pay for the same thing

	Alibaba	SoftBank
Instrument	New equity	Seven-year retail bonds
Size	HK$80bn (~$10.2bn)	¥1tn (~$6.3bn)
Price / coupon	HK$112.70 a share	4.3%-4.9% indicative
Use of proceeds	100% into full-stack AI	AI investment and refinancing
Share reaction, 24 Aug	−8.5% to HK$112.50	−5.3% to ¥4,975

One diluted its owners. One levered them. The market marked both down.

The Alibaba detail worth stopping on

The placement priced at HK$112.70. The stock closed Monday at HK$112.50 — beneath the price the new investors paid.

That does not happen in a well-absorbed deal. It means the buyers of $10.2 billion of new stock were, within a session, holding a loss, and it tells you the size cleared through price rather than through demand.

Alibaba was explicit that 100% of the proceeds go into AI infrastructure. So the market was asked to fund a capex programme with no revenue attached yet, and it agreed — at a discount, and grudgingly. The stock recovered to HK$115.50 by Thursday, which is a partial answer, not a verdict.

The SoftBank detail worth stopping on

A 4.3% to 4.9% coupon on seven-year paper, sold to Japanese retail investors, for the largest retail bond in the country's history.

Japanese households are being offered a yield that did not exist domestically for a generation, in exchange for financing an AI portfolio whose returns arrive well beyond the seven-year tenor. The bond matures long before most of what it funds is expected to pay.

That mismatch is the trade. SoftBank has run it before, successfully and unsuccessfully. What is different now is the price of the money: the previous record issue in April 2025 was ¥600 billion, and rates were a different animal.

Why the timing is the story

These two raises priced in the same week that Fed chairman Kevin Warsh moved September rate-hike odds to a coin flip, and the same week Nvidia guided to $108 billion of quarterly revenue.

Put those together and the picture is uncomfortable. Nvidia's revenue is somebody's capital budget. The capital budget is increasingly funded with issued paper — equity where the stock will absorb it, debt where it will not. And the cost of that paper is being repriced upward while the projects it funds are dated years out.

AI capex has quietly become a financing story rather than a demand story. Demand is not in question; Nvidia settled that on Wednesday. What is in question is who funds it and at what cost.

How I read it

Watching who has to issue and on what terms is a better read on this cycle than watching order books.

A hyperscaler funding data centres from operating cash flow is in a fundamentally different position from a company placing stock below the last traded price or selling seven-year retail paper at 4.9%. The first can slow down voluntarily. The second has a maturity schedule.

What I would watch: SoftBank's final coupon on 4 September. The indicative range was set before Warsh spoke. If it prices at the top of the range or above it, that is the first hard number on what a September repricing costs the AI build-out — and it will be a real number, not a survey.

More market analysis by Ruslan Averin at averin.com.

The Most Crushing Sanctions Ever Announced, and Oil Fell 6.5%

Ruslan Averin — Fri, 28 Aug 2026 22:45:38 +0000

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

On Monday the administration announced a global sanctions plan against Iran, with penalties for any country helping Tehran evade them, framed as economic warfare on an unprecedented scale.

Brent fell 2.4% that day. It fell another 3.9% on Tuesday to an eleven-day low. It ended the week at $88.29 against $94.39 the previous Friday — down 6.5%.

The week in one column

Day	Brent close	Move
Fri 21 Aug	$94.39	—
Mon 24 Aug (sanctions announced)	$92.17	−2.4%
Tue 25 Aug	$88.58	−3.9%
Wed 26 Aug	$87.84	−0.8%
Thu 27 Aug	$89.70	+2.1%
Fri 28 Aug	$88.29	−1.6%

Two weeks ago Brent printed $91.36 on an expired deadline and a closed strait. This week it took the largest sanctions announcement of the conflict and gave up six dollars.

Why an oil trader sells that headline

Sanctions are not a supply event in the way a blockade is. They are a routing event.

Barrels under sanction do not stop existing. They move through intermediaries, get discounted, take longer voyages and end up in buyers who accept the paperwork risk for the price. Physical supply to the world falls by considerably less than the political language implies, and the market has fifteen years of practice pricing exactly this.

There is a second, sharper reading, and I think it is the dominant one. Announcing maximum economic pressure is a statement about which instrument is being used. A government preparing to strike does not lead with a sanctions rollout. So the announcement that sounds most aggressive is also, to a trader, the clearest signal that the military option has been deferred — and it is the military option that the risk premium was pricing.

The premium in the price was never about Iranian export volumes. It was about the Strait of Hormuz. Monday's announcement quietly said the strait is not the near-term battleground, and $6 came out.

The consequence that matters outside energy

Two weeks ago I wrote that the question for the long end of the Treasury curve was whether the next leg higher in yields arrived with oil or without it. This week gave the answer in the cleanest possible form: Brent fell 6.5% and the front end of the curve repriced toward a hike anyway.

That matters more than the oil price itself. It means the inflation worry Kevin Warsh voiced at Jackson Hole cannot be dismissed as an energy pass-through. Energy went the other way and he was still hawkish.

An inflation problem with an oil cause has a mechanical cure. An inflation problem without one requires a policy rate.

How I read it

I am treating the risk premium as largely discharged rather than absent. Hormuz has not been resolved; it has been de-prioritised, and de-prioritisation reverses in a single headline.

What I would watch: the spread between Brent and WTI. If the strait genuinely stops being a live risk, the spread compresses toward the freight differential and stays there. If it stays wide while the flat price falls, the market is still paying for the option on escalation — and the option, not the barrel, is where this trade is decided.

More market analysis by Ruslan Averin at averin.com.

Meta Paid $16.7 Billion and the Stock Went Up

Ruslan Averin — Fri, 28 Aug 2026 22:45:08 +0000

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

Meta agreed to pay $16.68 billion to settle claims from 29 state attorneys general that it built Facebook and Instagram to be addictive to children. The judge approved it the same afternoon.

The stock closed up 1.1% that day, and up 5.1% on the week.

What the market was pricing

A settlement of this size is not a rounding error even for Meta. The market took it as good news anyway, for a reason that is worth stating precisely.

An open trial is an unbounded distribution. Discovery produces documents, testimony produces headlines, a verdict produces a number nobody can model, and an adverse judgment produces a precedent that other plaintiffs use. A settlement converts all of that into one line: $16.68 billion, known, paid, finished.

Markets pay for the conversion of an unbounded liability into a bounded one, and they will pay a premium for it. That premium exceeded the cash on Wednesday.

The part that is not cash

The operational terms matter more than the cheque, and they are easy to skip past.

Meta must impose daily time caps on teen accounts, block overnight access, strengthen age verification to keep children off the platforms entirely, and expand parental controls. Nationwide.

Every one of those reduces engagement in a cohort. Time caps directly cut sessions. Overnight restrictions remove hours. Real age verification removes users who should not have been there and who were, until now, monetised.

The $16.68 billion is paid once. The engagement terms compound, quarter after quarter, in the segment that shapes the next decade of habit formation.

Why the total could reach $17.1 billion

Several attorneys general put the ceiling higher, contingent on other platforms — TikTok and YouTube among them — settling their own cases with the states.

That structure tells you this was never really about one company. It is the establishment of a category-wide standard, and Meta went first, which historically means it set the template and paid to define it. Whoever settles next negotiates against these terms.

How I read it

The headline number is the least interesting figure in the release.

What I would model instead is teen daily active time under mandatory caps and overnight blackouts, and the size of the cohort that fails a strengthened age check. Neither has ever been disclosed cleanly, which is precisely why the market defaulted to pricing the cash it could see.

What I would watch: the first quarterly report after implementation, and specifically whether Meta discloses any engagement metric broken out by age. If it does not — and I expect it will not — the market will keep marking this as a solved problem, and the cost will surface slowly, in a growth rate rather than in a settlement line.

More market analysis by Ruslan Averin at averin.com.

The $53 Billion Buyout Died and PayPal Lost a Year of Gains in a Day

Ruslan Averin — Fri, 28 Aug 2026 22:44:37 +0000

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

Advent and Stripe walked away from PayPal. The bid had been $60.50 a share — more than $53 billion, and one of the largest leveraged buyouts anyone has attempted.

PayPal closed Friday at $53.66, down 12.7% on the day, having been 16% lower in the premarket.

The round trip

Date	PYPL close
Week of 6 July, pre-bid	$46.32
Week of 13 July, reports emerge	$56.56
Thu 27 Aug	$61.47
Fri 28 Aug	$53.66

Read the first and last rows. The stock is still 16% above its pre-bid level. The market removed most of the premium on Friday and deliberately left some of it on the table.

That residual is the interesting number. It is either a bet that another buyer appears, or a re-rating of the standalone business after a quarter that beat most estimates. Those are very different theses trading at the same price, which is usually where mistakes get made.

The reason the deal died is the week's real story

The board thought $60.50 was inadequate. The consortium faced regulatory and financing hurdles. And PayPal's own good quarter worked against the transaction — a 40% run in the stock raises the price the buyers must pay while the debt they would use to pay it becomes more expensive.

That last mechanism deserves attention beyond this one deal. A leveraged buyout of $53 billion is an interest-rate instrument as much as a corporate one. It works if the cost of debt over the holding period is roughly what you underwrote.

On the same day this deal collapsed, the Fed chairman told Jackson Hole he still has work to do on inflation, and the market moved September rate-hike odds to a coin flip. Sponsors underwriting seven-year paper into a curve that is repricing upward have a defensible reason to stop, and a target whose price has already run gives them a graceful exit.

What holders are actually left with

Three things worth separating.

A business that beat estimates. The quarter that helped kill the deal was a real quarter.

No floor. Deal speculation was acting as a soft bid under the stock for six weeks. That is gone, and the marginal buyer now has to want PayPal on its own numbers.

A signal about strategic interest. Stripe examined its largest listed competitor and concluded it would not pay $60.50. That is information about how a well-informed operator values these assets.

How I read it

I do not own deal spreads and this week is a reminder why. The event risk is binary, the downside arrives in one session, and the thing that broke the trade — the cost of financing — had nothing to do with either company.

The more useful reading is as a marker. If a $53 billion LBO cannot get financed in August 2026, the large-cap take-private pipeline is narrower than the headlines of the past year suggested, and any stock carrying a speculative buyout premium is carrying the same risk PayPal just realised.

What I would watch: whether another sponsor appears at a lower price in the next quarter. If nobody does, the 16% still sitting above the pre-bid level has to be justified by earnings rather than by hope — and the first quarter that disappoints will collect it.

More market analysis by Ruslan Averin at averin.com.