Peakhill in 2025: The Capital Cushion Looks Thick, but the Real Test Is the CMHC Exit Path

Getting To Know The Company

Peakhill is not a generic non-bank lender spread across many unrelated credit pockets. It is a limited partnership built to fund a relatively short-duration mortgage book, mainly Canadian multifamily real estate, and service its bonds through fast recycling of bridge loans into permanent CMHC financing. In other words, the economics here rely less on holding long-dated credit for years and more on underwriting well, originating a short bridge loan, and exiting on time through a government-backed takeout path that effectively defines the whole model.

What is working right now is clear enough. The portfolio nearly doubled in a year, revenue rose to CAD 41.5 million, net income reached CAD 26.4 million, and the partnership was already back in the bond market in January 2026 with a series expansion. At the bondholder level, there is also a reasonably comfortable cushion: covenant headroom is wide, interest coverage stood at 4.89, and most of the book is short enough to turn over quickly if the exit channel continues to work.

But a surface read can miss the active bottleneck. Peakhill presents a 60% capital ratio, which sounds highly conservative. In practice, that figure includes a subordinated shareholder loan of CAD 91.1 million. Accounting equity by itself was CAD 268.4 million at year end 2025, roughly 45% of the balance sheet, not 60%. That is still a meaningful capital base, especially for a bond-only issuer, but it is a different picture: less pure accounting equity, more hybrid support that is very useful for covenants and ratings, and less clean if the reader is trying to understand what really sits underneath the debt.

The second point a reader may miss is the quality of the book. Across the life of the portfolio, no realized losses have been recorded, and that naturally creates comfort. At the same time, by the end of 2025 there were already 9 loans in payment arrears totaling CAD 27.2 million, the watchlist had climbed to CAD 56.3 million, and ECL allowances reached CAD 6.59 million. So the story has not broken, but it is no longer as clean as a headline of zero realized losses would suggest.

That is exactly why 2026 looks like a proof year here. Not a breakout year, because the jump has already happened. Not a stabilization year, because the cut is not clean yet. It is a year in which Peakhill has to show that its Israeli capital-markets model, CAD mortgage assets funded by NIS public debt, can keep growing without the rising watchlist, the dependence on CMHC, and the future possibility of bank leverage at the partnership level making the story materially less clean.

From a market-read perspective, it also matters that this is a bond-only issuer. There is no listed equity, no short-interest read, and no equity-screen actionability. The near-term reading runs through three different filters: access to debt markets, covenant performance, and confidence that the CMHC exit channel keeps functioning even as the book scales up.

Those two charts explain why Peakhill is interesting right now. On one side, the book is concentrated in a clear economic engine, CMHC bridge lending. On the other, CAD 415.8 million out of CAD 584.9 million is meant to mature by the end of 2026. That creates flexibility, but it also means the model cannot afford many exit failures.

Events And Triggers

The initial bond opened the tap, and the expansion tested market depth

In September 2025 Peakhill completed its Series A issuance at a principal amount of NIS 563.451 million and received roughly CAD 227.2 million net after costs. This was not just a financing event. It was the launch point of the entire Israeli public-markets model: a short Canadian mortgage book funded by public NIS debt, with FX hedging and a deed of trust built around minimum equity, interest coverage, and equity-to-assets covenants.

That money went to work quickly. The partnership itself says that after closing the transaction it increased its mortgage book by CAD 179.4 million. At the same time, gross mortgage investments rose during 2025 from CAD 267.7 million to CAD 584.9 million. In other words, the bond market did not merely strengthen an existing balance sheet. It financed an aggressive acceleration of the portfolio.

The second trigger came immediately after the balance-sheet date. In January 2026 the partnership expanded the series by another NIS 300 million principal amount, for gross proceeds of NIS 310.2 million, and the outstanding series rose to NIS 863.451 million principal amount. That is a positive signal because it shows the issuer could come back to market quickly. But there is also a practical yellow flag here: the auction received no public orders at all, and all 300,000 units were allocated based on pre-commitments from classified investors. So capital-markets access exists, but for now it rests on institutions rather than broad market demand.

The rating stayed stable, but the more important signal is who actually bought

Midroog reaffirmed the issuer and the series at A2.il with a stable outlook in January 2026. That matters, because in a structure like this the rating is not just a label, it is part of the market-access equation. But for the forward read, the sharper signal is not the rating itself but the fact that the series expansion priced without any public demand. That does not mean the market is shut. It does mean the trust test currently sits inside a narrower, more professional investor base.

The next chapter is already in the filings, even if not yet in the balance sheet

The third trigger is one a casual reader can skip because it is not yet in the 2025 numbers. At group level, Peakhill Capital is in the process of consolidating three existing credit lines into one CAD 275 million facility. As of year end 2025, that bank funding is outside the partnership and explicitly non-recourse to it and to the bondholders. That supports the quality of the structure today. But the company also says explicitly that such facilities could in the future be borrowed through the partnership and recognized in its financial statements. That means today’s cleaner issuer-level balance sheet should not be read as a permanent fixed point.

That chart tells the story better than a long narrative. In 2024 this was almost a debt-free issuer at the partnership level. In 2025 it became a full public bond issuer. That is not negative by itself. It is the point of the story. But it is also the moment when the reader needs to stop treating the business like a private fund and start reading it like a public issuer that lives on a precise balance between portfolio growth, FX hedging, and refinancing ability.

Efficiency, Profitability, And Competition

2025 growth came from volume, not from stronger pricing

Interest income rose 48% to CAD 36.0 million, and total revenue rose 56.6% to CAD 41.5 million. But the more interesting point is that the growth did not come from a higher-yielding book. It came despite lower portfolio yield. In the mortgage-investment note, the weighted average interest rate of the portfolio fell from 11.31% at the end of 2024 to 8.60% at the end of 2025. So Peakhill grew through scale and portfolio mix, not by pushing price harder.

That makes economic sense. As the book shifts more toward CMHC bridge lending and a broader short-duration portfolio with a more defined exit path, yield can come down relative to a smaller or more opportunistic book. The right question, then, is not whether yield declined. It is whether that decline bought higher quality. The answer for now is only partial. On one side, the book is larger, more diversified, and shorter. On the other, the watchlist and the amount in arrears both rose, so the reader cannot yet conclude that lower yield fully purchased better quality.

Not every revenue line was generated directly by the mortgage book

There is an important earnings-quality point here. Out of CAD 5.478 million of other income in 2025, CAD 4.8 million came from commitment fees earned from Peakhill Capital. In other words, roughly 88% of non-interest income came from a related-party relationship. That is not improper. It does mean that the seemingly diversified revenue mix is much less diversified in practice.

At the same time, on the expense side, management fees declined from CAD 1.853 million to CAD 1.488 million, with the partnership explaining that no management fees were paid starting in October 2025. So 2025 profit includes help from a change in the internal fee layer, not only from cleaner operating economics.

The competitive edge is real, but it also creates concentration

Peakhill does have a real competitive angle in Canada. It operates in a segment where banks tend to be less flexible on mid-sized commercial lending, and it builds underwriting around the CMHC takeout path from day one. That allows it to originate a bridge loan with a largely pre-defined exit. When it works, it is a strong engine: borrowers get speed and flexibility, and the partnership gets short credit with a relatively visible exit route.

But that same edge also creates the main concentration point. When 71.3% of the portfolio is made up of bridge loans eligible for permanent CMHC financing, the key competitive advantage and the key operating risk become the same thing. If CMHC remains efficient and available, the book works well. If the exit pipe slows, partially closes, or becomes more selective, the same advantage can quickly turn into a liquidity and credit-quality pressure point.

That chart sharpens another important point. 2025 is the first year in which the public-debt layer truly sits inside the income statement. Up to 2024, earnings rose almost in lockstep with revenue and without a heavy financing burden. In 2025, CAD 6.8 million of net financing expense was added, and net income still rose. That is a positive sign for debt-carrying capacity. But it is still only year one of this structure. To read it as a durable base, the market needs more reported periods.

Cash Flow, Debt, And Capital Structure

Cash flow, in all-in cash-flexibility terms

For a non-bank lender, negative operating cash flow does not automatically mean distress. Here it mainly means the book grew quickly, because new loan originations are classified inside operating activity. That is why the cash framing has to be precise.

In all-in cash-flexibility terms, meaning after the period’s actual cash uses, the 2025 picture looks like this: the partnership started the year with CAD 25.5 million of cash, used CAD 288.7 million in operating activity, received CAD 270.0 million from financing activity, and ended the year with CAD 6.8 million. That is not a collapse story, because book growth is the core activity. But it is also not a picture of abundant spare liquidity.

Against that, working capital actually looks strong at CAD 340.4 million, because most of the book is current and short. So two pictures need to be held together at once. The first says the cash balance itself is not large. The second says the asset book is relatively liquid as long as the exit path keeps functioning. Anyone looking only at the cash line will think the risk is too high. Anyone looking only at working capital will miss how dependent the structure is on fast turnover.

The debt layer, and the gap between accounting equity and covenant equity

This is probably the single most important accounting point in the filing. At the end of 2025 Peakhill carried CAD 234.5 million of bonds, CAD 91.1 million of subordinated shareholder debt, and CAD 268.4 million of partners’ equity. The deed of trust does not measure equity only through accounting equity. It measures it after adding subordinated shareholder loans. That is how the 60% capital ratio is created.

Put more directly, for bondholders this is a supportive structure: the shareholder loan sits below them, carries no interest, and improves covenant headroom. But in terms of capital quality, the distinction matters. A meaningful part of the apparent cushion is not pure accounting equity but subordinated insider debt. It certainly strengthens protection for bondholders, but it is not identical to permanent, non-debt capital.

The nuance deepens because the shareholder loan is also described as debt that is payable on demand subject to covenant conditions. That does not make it an immediate problem. It does mean the 60% capital headline needs a clear footnote. The reader should not reject the figure. The reader should understand exactly what sits inside it.

FX and hedging

Peakhill’s debt is denominated in shekels while its assets sit in Canadian dollars. That is a real cross-currency exposure, so it is constructive to see an actual hedging layer in place. As of year end 2025 the partnership reported a derivative asset of CAD 7.646 million and disclosed three forward contracts covering the March and September 2026 payments and the original series principal through September 2026. The January 2026 series expansion was also hedged after the balance-sheet date. That materially reduces short-term currency risk, but it does not eliminate it over time, because the hedge program itself requires rolling and operational discipline.

The positive side: covenant headroom is still wide

The reassuring part of the story is that covenant headroom still looks comfortable at the end of 2025. Covenant equity stood at CAD 359.5 million against a CAD 220 million minimum for the interest-rate adjustment test, the equity-to-assets ratio stood near 60% against a 28% floor, and interest coverage was 4.89 against a 1.25 minimum. This is not a near-edge issuer. The correct read is not immediate balance-sheet stress. It is a structure that still looks comfortable, but one that will have to defend that comfort if growth continues while the book stays short and exit-dependent.

Outlook

Before getting into the actual outlook, four points need to stay front of mind:

Point one: the 60% capital ratio is a real covenant number, but not pure accounting equity. Anyone reading it that way is missing the actual support structure.

Point two: the CAD 27.2 million arrears bucket at the end of 2025 is almost as large as the cumulative lifetime default volume Peakhill has disclosed to date, CAD 30.9 million.

Point three: more than two-thirds of the book is meant to turn within a year, so 2026 will be judged first on refinancing performance, not on portfolio size alone.

Point four: the January 2026 series expansion proves market access, but it also shows that access is still narrow and institutionally driven.

From there the forward read becomes fairly clear. 2026 is not the year to ask whether Peakhill can grow. It has already shown it can originate loans, expand the book, and raise debt. The real question is whether it can grow without the core stress points of the model, CMHC exits, watchlist behavior, hedge discipline, and capital-structure quality, starting to weigh on the story.

The constructive scenario is that Peakhill uses the January 2026 proceeds to keep expanding the book while at the same time working down the watchlist without realized losses. If that happens, 2025 will look in hindsight like the transition year in which the issuer moved from launch mode to proof mode.

The less comfortable scenario is that the book keeps growing, but CMHC exit timing stretches, the arrears bucket does not come down, and the group chooses to bring bank leverage into the partnership to keep pushing growth. In that case Peakhill could still remain inside covenants, but the quality of the story would change. It would shift from a relatively simple bond-funded short mortgage book into a denser structure of bonds, subordinated shareholder support, derivatives, and potentially bank funding.

That chart shows why 2026 is a proof year. Loans in arrears rose only moderately as a share of the book, from 4.1% to 4.6%, but the broader watchlist jumped from 5.7% to 9.6%. In other words, before loans visibly break, more and more of the book already requires special attention. That is usually the point at which the analyst needs to slow down before extrapolating a smooth growth story.

What has to happen over the next 2 to 4 quarters for the picture to improve is fairly straightforward. First, a material portion of the CAD 415.8 million maturing by the end of 2026 needs to exit on time. Second, the 9 loans already in arrears need to resolve without turning the zero-realized-loss record into an actual capital hit. Third, the January 2026 bond expansion needs to turn into income without creating a new funding problem. And fourth, if bank funding is brought into the partnership, the company will have to explain clearly why that improves economics rather than just pushing growth further out.

Risks

The first and most important risk is exit-path risk. Peakhill works with a very clear model: bridge now, refinance into CMHC later. That is its strength, but it is also its main dependency. When more than 90% of historical loans have exited through refinancing, and when 71.3% of the current portfolio is built around that same path, any slowdown in that channel can first show up in the watchlist, then in arrears, and then in realizations.

The second risk is credit quality weakening underneath a zero-loss headline. No realized losses have been recorded, but the ECL allowance has already risen to CAD 6.59 million, and the partnership’s own auditors treat this estimate as a key audit matter. In addition, a move of only 10 basis points in the macro factor used in the ECL model changes the provision by CAD 653 thousand. That number does not break a platform like this on its own, but it is a reminder that reported earnings here are sensitive to judgment and environment, not only to hard cash outcomes.

The third risk is a related-party dependent structure. A large part of other income comes from a related party, management and servicing fees are paid to a related party, and the underlying assets were transferred from an affiliated transferring entity. This is not abnormal for this kind of issuer, but it does mean the investor needs to watch earnings quality, not just earnings size.

The fourth risk is a possible shift toward a more layered funding structure. Today, bank funding remains outside the partnership and non-recourse to it. That is supportive. But the fact that the company explicitly says a consolidated CAD 275 million facility could in the future be borrowed through the partnership means the question is already live. If that happens, the story will no longer be only bonds against a short mortgage book.

The fifth risk is FX and hedge execution risk. For now the hedge program looks orderly, but the liabilities are in NIS and the assets in CAD. Every new series expansion requires the company to keep rolling the hedge discipline without operational misses, otherwise earnings and capital can move for reasons that have nothing to do with credit quality.

Conclusions

Peakhill reaches the end of 2025 with two stories living side by side. The first is constructive: a platform that can scale, proven access to the debt market, a relatively short book, and comfortable covenant headroom. The second is much more cautious: the capital cushion is less pure than the 60% headline suggests, the watchlist jumped, and the CMHC exit model has become not only the key advantage but also the key test.

Current thesis: Peakhill still looks like a debt issuer with a mortgage book that is functioning, but the quality of the story now depends much more on working down the watchlist and maintaining CMHC refinancing than on printing another bond deal or adding another leg of portfolio growth.

What changed relative to the older understanding of the company? First, it is no longer an almost debt-free partnership-level entity. It is now a full public bond issuer. Second, book quality already requires work: loans in arrears are up, and the broader watchlist is up even faster. Third, the Israeli debt market has shown that it is willing to fund the story, but at this stage mainly through institutional money.

The strongest counter-thesis is that Peakhill has already shown, over years, that it can resolve defaults without realized loss, that it runs a diversified book, that it benefits from strict CMHC underwriting discipline, and that it ended 2025 with 4.89 interest coverage and wide covenant headroom. That is a serious argument and deserves respect. But as of the end of 2025 it is still an argument that needs further proof, because the scale of the book and the public-debt layer have grown faster than the historical track record supporting them.

What could change the market’s interpretation over the short to medium term? Mainly two things: a quiet resolution of the arrears bucket over the next few quarters, or the first sign that bank funding needs to enter the partnership to sustain growth. The first would materially strengthen the thesis. The second would not necessarily break it, but it would change its character.

Why this matters: Peakhill is building a bridge between Israeli public debt and short Canadian real-estate credit. As long as that bridge keeps working, it benefits from an efficient model, relatively modest leverage, and rapid scaling capacity. If the exit channel starts to slow, that same bridge becomes the place where risk begins to accumulate.

What has to happen next is fairly clear. Over the next 2 to 4 quarters the company needs to show that loans continue to exit on time, that the watchlist does not keep swelling, and that the next phase of growth does not come at the cost of a more complicated funding layer. What would weaken the thesis is some combination of a further rise in the watchlist, bank leverage moving into the partnership, or the first real crack in the no-realized-loss record.