管理層發言
Good day, and thank you for standing by. Welcome to BXP's Q2 2026 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speakers' presentation, there will be a question-and-answer session. To ask a question during the session, you will need to press star 11 on your telephone keypad. You will then hear an automated message advising your hand is raised. To withdraw your question, please press star 11 again. We ask that you please limit your questions to no more than one, but feel free to go back into the queue. And if time permits, we will be happy to take your follow-up questions at that time. Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker, Helen Han, Vice President of Investor Relations. Please go ahead.
Good morning, and welcome to BXP's Second Quarter 2026 Earnings Conference Call. The press release and supplemental package were distributed last night and furnished on Form 8-K. In the supplemental package, BXP has reconciled all non-GAAP financial measures to the most directly comparable GAAP measure in accordance with Regulation G. If you did not receive a copy, these documents are available in the Investors section of our website at investors.bxp.com. The webcast of this call will be available for 12 months. At this time, we would like to inform you that certain statements made during this conference call which are not historical may constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act. Although BXP believes the expectations reflected in any forward-looking statements are based on reasonable assumptions, it can give no assurance that its expectations will be attained. Factors and risks that could cause actual results to differ materially from those expressed or implied by forward-looking statements are detailed in yesterday's press release and from time to time in BXP's filings with the SEC. BXP does not undertake a duty to update any forward-looking statements. I would like to welcome Owen Thomas, Chairman and Chief Executive Officer; Doug Linde, President; and Mike LaBelle, Chief Financial Officer. During the Q&A portion of our call, our regional management teams will be available to address any questions. We ask that those of you participating in the Q&A portion of the call please limit yourself to one question. If you have an additional query or follow-up, please feel free to rejoin the queue. I would now like to turn the call over to Owen Thomas for his formal remarks.
Thank you, Helen, and good morning to all of you. BXP delivered a very strong second quarter, both operationally and financially. FFO per share exceeded both our guidance and consensus estimates by $0.08 and we raised the midpoint of our 2026 FFO per share guidance by $0.05. We also made meaningful progress against the business plan we articulated at last year's investor conference. Leasing results were strong, in-service portfolio occupancy increased significantly, additional asset sales progressed, and our development pipeline was active with project deliveries, launches, leasing, and capital raising. Our first business plan priority is to lease space and improve portfolio occupancy. We had a great quarter completing nearly 1.8 million square feet of leasing—29% above our 10-year historical average for the second quarter. Year-to-date, we have leased over 3 million square feet and our in-service portfolio occupancy also rose materially for the third quarter in a row. This outcome reflects strong execution by our leasing teams as well as a very healthy environment for leasing premier workplaces. AI continues to be enormously beneficial to BXP's leasing activity. Our current and prospective clients are generally experiencing increasing earnings in an AI-powered U.S. economy, are more often expanding than contracting their space requirements and in many cases are also upgrading their space. We are leasing space to AI companies in San Francisco, New York, and Seattle, to companies displaced by growing AI firms, and to our core financial, legal, and business services clients that support the AI industry. While AI's long-term impacts remain difficult to predict, research shows that technology advances historically increase the share of office-based jobs. Additionally, AI will likely exert a greater impact on less adaptive back-office workers, and these roles make up a smaller share of employment in knowledge center gateway markets and in premier workplaces. Further, non-office-using remote jobs, which generally have more process and analytical than interpersonal requirements, will be more disrupted by AI. Lastly, companies winning in an AI-enabled economy will be more profitable and face more intense competition for talent, leading to less price-sensitive demand for easily commutable and desirable workplaces for their employees. For all these reasons, we believe premier workplaces located in gateway market knowledge centers are best positioned to benefit from—and at worst to be the most immune from—AI impacts on the labor force. As proof, the premier workplace segment of the office market where BXP is a clear leader continues to materially outperform the broader office market. Premier workplaces represent roughly the top 14% of space and 8% of buildings in the four CBD markets where BXP has a major presence. Direct vacancy for premier workplaces in these four markets is 8% versus 13.5% for the broader office market, while asking rents for premier workplaces continued to command a premium of more than 60% over the non-premier buildings. With an 8% vacancy rate, positive net absorption and limited new construction on the horizon, premier workplaces and BXP core markets are set up for material rent increases which have already commenced in many submarkets. Given these positive market forces, we are well on our way to accomplishing our two percentage point occupancy gain goal in 2026, reinforcing our confidence that our target of four percentage points of total occupancy improvement over 2026–2027 remains very much on track. Our second business plan goal is to raise capital and optimize our portfolio through asset sales. At our investor conference, we communicated an objective to generate in aggregate $1.9 billion in net sale proceeds by 2028 from the sale of land, residential, and non-strategic office assets. We continue to make progress in the second quarter and are well ahead of schedule. We have raised $370 million in total net sale proceeds so far this year and more than $1.2 billion since our investor conference. In addition, we have six assets under contract for sale with total net proceeds of approximately $240 million, $180 million of which is scheduled to close in 2026. Two of the assets currently under contract for sale are office buildings in Washington, D.C., which are scheduled to close this quarter. We are also in various stages of marketing several additional assets, including 7 Times Square in New York City. As of now, future net proceeds from dispositions possible in 2026 could aggregate up to an additional $500 million bringing our total net proceeds from asset sales to $1.7 billion by year-end, and we continue to explore additional capital raising opportunities. Supporting our disposition efforts, office volume in the private markets remains reasonably healthy with financing available at scale, particularly in the CMBS market. In the second quarter, significant office sales were $12.6 billion, down 13% from the first quarter and essentially flat from the second quarter of 2025. Though there continue to be very few premier workplace assets trading, there were a couple of transactions in the quarter with relevance to BXP's portfolio. One, Marina Park Drive located in the Seaport of Boston is under agreement to sell for approximately $435 million which represents pricing of nearly $900 per square foot and an initial cap rate in the low-7% range. The asset comprises 495,000 square feet, is 99% leased with above-market rents, and is being sold by an adviser to the operating arm of a non-U.S. pension plan. Further, Tower 1 at West Main located in Bellevue, Washington is under agreement to sell for approximately $340 million, representing pricing of around $930 per square foot and a 6.75% initial cap rate. The 365,000-square-foot building is fully leased to Amazon on a long-term basis and was sold by a local developer to an adviser. BXP's third business plan goal is to grow FFO through new developments selectively with office given market conditions and more actively for multifamily with an equity partner. For office, we have and expect to allocate more capital to developments than acquisitions due to the materially higher yields available. This quarter, we delivered into service 290 Binney Street, a 570,000-square-foot lab building fully leased to AstraZeneca located in the life science nexus of East Cambridge. The project is a great example of BXP's development skills: creating value for shareholders where we establish development rights through executing a complex infrastructure enhancement. We fully leased the asset before commencement, sold a 45% stake in the property at a profit to a financial partner, and delivered the project $20 million below budget and two months ahead of schedule. BXP's $488 million investment for its share of the project is yielding an 8.9% unleveraged cash return and a 10.3% GAAP return. BXP's largest development underway is 343 Madison Avenue, our premier workplace tower in New York City with direct access to Grand Central Terminal. This past quarter, we signed a 148,000-square-foot lease with McDermott Will & Emery at the bottom of the high-rise bank of the building, and Starr expanded by two floors in the mid-rise, bringing us to 50% leased. Further, we are in lease negotiations with a two-floor client in the podium, which if completed, would bring us to 56% leased. Lastly, we are exchanging proposals with another client requiring five floors at the base of the podium, which would bring the project to nearly 70% leased. Though we have received single-floor inquiries for the seven floors remaining at the top of the building, we expect continued rent appreciation and will likely lease these floors closer to delivery given their ability to command market-leading rents. We have procured 94% of the construction costs on budget. Leasing economics have been at or above forecast and our projections remain on track for a stabilized unleveraged cash return of 7.5%–8% upon delivery in 2029. Yesterday, we closed a 60% loan-to-cost $1.2 billion construction loan for the project on attractive terms and have a letter of intent with an equity partner for an $80 million investment representing a 10% interest in the project, with a basis above our costs. We expect the equity investment to close this quarter and our marketing efforts continue with the goal of ultimately monetizing a total of 30%–50% of the project over time. The value of the development continues to rise as we lease space and get closer to delivery. This past quarter, we launched the development of our Worldgate multifamily project comprising 359 wood-frame residential units located in Herndon, Virginia. The project's budgeted cost is $132 million, and we have secured a financial partner to supply 80% of the equity as well as the construction financing. BXP originally bought into the Worldgate property, which comprised an empty office building and parking garage on 10 acres in 2023. The project was rezoned for residential. The for-sale component is under contract for sale to a homebuilder, and the apartment development will entail demolishing the office building and utilizing structured parking. BXP will earn a profit from the total monetization of our investment in Worldgate, and has reinvested our share of the proceeds from the contribution of the apartment land back into the development joint venture for a 20% interest. We have additional residential projects in Weston, Massachusetts and Santa Monica, California, that we intend to launch next year. This past quarter, we also signed a 320,000-square-foot long-term lease with Boston Dynamics, which will create a state-of-the-art robotics and AI center at Reservoir Place, a 360,000-square-foot office building BXP had taken out of service in Waltham. BXP will invest $87 million in the building and expects to earn an initial cash return of over 10% including an inferred value for the existing improvements. The project is expected to be delivered into service in the second quarter of next year. BXP's current development pipeline, comprising seven office and residential projects underway totaling 3.5 million square feet and $3.2 billion of BXP investment, continues to deliver external growth over the longer term. In conclusion, BXP is set up well for success. New construction for office has virtually halted, already leading to higher occupancy and rent growth in most submarkets where BXP operates. Debt capital is readily available for premier workplaces at attractive credit spreads. BXP continues to capture market share driven by our stability, reliable client service, and a lighter competitive landscape across many markets. BXP remains comfortably on track with our business plan, which, if successful, will lead to increasing portfolio occupancy and FFO per share, deleveraging, external growth from development, and a more AI-enabled gateway CBD premier workplace-concentrated portfolio in the years ahead. Over to Doug.
Thanks, Owen. Morning, everybody. Owen did a great job articulating our view on why AI is critically important to the demand picture. Equally important, as a public company, the rhetoric and conjecture around the impact of new AI technology on the future of office-using jobs has become much more balanced and constructive since February of this year. In each of our markets, our portfolio has seen a pickup in demand. In our best markets, that demand is coming from clients that are expanding across a wide spectrum of industries, including technology, AI, defense and cybersecurity, asset management, financial services, and professional services. In our other markets, the demand is due to decisions around upgrading space or changes in geographic preference as our clients look to maximize the desirability of their space for their associates. It is all encouraging for the premier office product. BXP had great top-line revenue results this quarter, and I want to focus my time on the improvements in our occupancy, which drove much of that outperformance. In June, when we were with you at Nareit, we said our leasing progress was ahead of schedule relative to our anticipated occupancy pickup. We ended 2025 at 86.7% occupied. We finished the first quarter at 87.4%. And as of June 30, 2026, we are 88.4% occupied. So we have gained 170 of the 200 basis points that we originally expected for 2026. We had guided to an average occupancy during the year of 88.2% and we are ahead of plan. While the individual transactions may be granular, the simple explanation is that we leased space more quickly than we expected. Most importantly, we continue to lease vacant and near-term expiring space. In the first quarter, BXP's total leasing volume was 1.14 million square feet, and we executed leases on 700,000 square feet of vacant space. In the second quarter, we completed 1.76 million square feet and covered an additional 380,000 square feet of vacant space and renewed or backfilled 600,000 square feet of 2026 and 2027 expirations. 190,000 square feet of our activity this quarter was at 343 Madison. As Owen mentioned, 322,000 square feet was with Boston Dynamics at Reservoir Place. All vacant space, but those are not in-service properties. We start the third quarter with a signed but not occupied portfolio of about 1.3 million square feet with 1.1 million expected to commence in 2026. The remaining calendar year 2026 known expirations are down to 300,000 square feet. This means we are going to pick up 800,000 square feet of occupancy or another 170 basis points and close the year closer to 90% than 89%. Our 2027 expirations currently stand at 1.77 million square feet. We have known vacates of about 1 million and have good clarity on about 550,000 square feet of either renewals or replacement tenants for those expirations. We also have 250,000 square feet of signed leases that we expect to commence in 2027. Our pipeline of leases either executed or in negotiation after the second quarter stands at 1.3 million with about 350,000 square feet of that involving vacant space. In addition, our active discussions are approaching 1.7 million square feet, which could impact another 450,000 square feet of current vacancy. Midtown Manhattan properties were up 14% in leasing. In San Francisco, 40% of the square footage in the statistics this quarter was in Mountain View, where new leases reset at rents of about $45 triple-net. And in Seattle, 70% of the square footage came from a low-cost expansion with a technology company at Madison Center, meaning very little tenant improvement expense. This quarter, we executed 21 leases over 20,000 square feet in the in-service portfolio; 48% of the square footage was renewals, extensions, or expansions, and 52% was with new clients. Existing client expansions encompass 275,000 square feet of that activity, and we had about 50,000 square feet of current client contraction. In the BXP portfolio, Midtown Manhattan, the Back Bay of Boston, and Reston, Virginia continue to have the tightest supply and therefore the most landlord-favorable market conditions. While San Francisco and Manhattan are dominating the landscape with technology and AI demand, we are also seeing demand elsewhere. We completed about 170,000 square feet of leasing in our Back Bay portfolio and are starting to see our first wave of renewals at 888 Boylston Street. The highlights of this quarter in the Boston region were the 322,000-square-foot lease with Boston Dynamics, which illustrates the point about AI leading to increased demand. This facility will house Boston Dynamics' advanced robotics and AI center and they announced the expected hiring of over 1,000 new employees. In our urban edge portfolio, demand around lab space remains lackluster. While the life science capital markets are very active with a series of Boston area IPOs and several big pharma acquisitions of Boston-bred biotechs, capital raising around the startup sector continues to be slow. It is the Series B, C, D companies that eventually move out of incubators into proprietary space that are still missing in the market. We continue to make progress at our Quarry asset, our largest availability in the Urban Edge, where we are in lease with a 50,000-square-foot client, another life science company, that is building 100% office space in our facility. In New York, at 360 Park Avenue South, we are in lease for the last floor from an expanding AI tech company, which will bring the building to 100% occupied. This quarter, we completed an extension and expansion with Rogo, a client that develops AI tools specifically for financial institutions and also announced job expansions. Across Madison Park at 200 Fifth, we are in lease for the remaining available space, and when complete, we will be 100% leased there as well. These two assets had almost 750,000 square feet of available space at the end of the first quarter of 2025. Our activity north of 42nd Street in Midtown this quarter also included expansions from financial advisers, asset managers, and law firms totaling 100,000 square feet. We also completed 10 transactions in Princeton totaling over 100,000 square feet. In San Francisco, the most significant momentum in our portfolio continues to be at 680 Folsom and 50 Hawthorne. During the quarter, we executed a 63,000-square-foot lease, and we are in discussions now with an applied AI company for a 35,000-square-foot floor and with an existing AI client about expanding into the final available floor at 680 Folsom. We have also had success with smaller technology companies expanding at 535 Mission and at Embarcadero Center. We recently completed two transactions and are in discussions with three more. We are approaching our first significant initial lease-up expirations at Salesforce Tower in 2027. Current market rents are 30%–40% higher than the expire rents in the building and still would be a significant discount to new construction economics. It is hard to find holes in the San Francisco demand picture when you have had 3 million square feet of positive absorption over the last two quarters. However, the one soft spot continues to be incremental demand growth from traditional financial services, professional services, and legal firms. In Mountain View, we completed 190,000 square feet of leases. Vacant space made up 50% of this activity, and we are in discussions with new clients for another 70,000 square feet of vacancy in the park. In Seattle, we completed over 100,000 square feet of leasing on vacant space this quarter, including a 44,000-square-foot expansion by Stripe and another floor with an AI company that expects to grow its headcount 4x in 2026. Activity in D.C. this quarter was concentrated in Reston where we leased over 125,000 square feet of 2027 expiring leases to defense contractors, cybersecurity firms, and a financial firm. In the District, we are in negotiations to lease 100% of the space that McDermott will be vacating at 500 North Capitol in late 2028 when we deliver 32. With the expected sale of two office assets, we are shrinking our District portfolio prior to adding our newly leased developments. In the interim, the D.C. team continues to field inbound requests from law firms that want us to identify sites and develop new projects like what we have achieved at 725 12th and 2100 M Street. We are working with an institutional owner to organize a joint venture on a third of these projects, and hope to have a lease commitment before the end of 2026. In summary, our assets are seeing strong demand growth. We are leasing space more quickly, and as Mike described, it is impacting our bottom line.
Thanks, Doug. Good morning, everybody. Today, I will cover our financing activities as well as our strong results for the second quarter earnings and provide an update of our full year 2026 earnings guidance. As Owen mentioned, we closed the $1.2 billion five-year construction loan to fund approximately 60% of the development cost of our 343 Madison project. The loan was competitively bid and we experienced strong demand from our largest banking partners. The demand allowed us to achieve very attractive pricing and terms relative to recent deals in the office construction loan market and it demonstrates the engagement of institutional lenders to finance premier quality office projects with our strong sponsorship. The pricing is floating at SOFR plus 250 basis points, with a reduction to 225 basis points upon the achievement of project milestones. The interest expense will be capitalized into the project cost and will not be included in our interest expense until completion in 2029. This is an important milestone for 343 Madison and it provides us with an additional capital source and financial flexibility. We are also focused on the upcoming refinancing of a $1 billion unsecured bond that carries a GAAP interest rate of 3.5% and expires this October. While rates markets have been volatile, the bond market has been very active with credit spreads near all-time tights. Our 10-year credit spreads are trading in the low 100s; if we were to issue a new bond today, it would likely price around 6% based on the current 10-year Treasury rate. With the success of our asset sales program and the financing of 343 Madison, we may elect to use available cash to reduce the size of this refinancing by up to $300 million to minimize dilution. We also continue to evaluate all the refinancing alternatives available as we seek to optimize our debt capital structure and mitigate the impact of the elevated interest rate environment. Now I would like to turn to our second quarter earnings results. We had a very strong quarter and reported FFO of $1.78 per share, which exceeded the midpoint of our guidance and consensus by $0.08 per share. Importantly, nearly all of our outperformance came from better results in portfolio NOI. Our revenues exceeded our expectations by $0.04 per share comprised of $0.03 per share of higher rental revenues and $0.01 per share of higher service income. Robust leasing activity drove higher rental revenue and occupancy this quarter. The revenue lift reflects earlier-than-anticipated occupancy and I do not expect it to compound into future projections. As Doug described, our leasing activity has beaten our expectations, with occupancy climbing by 100 basis points to 88.4% this quarter. We have increased our expectations for average occupancy for the year by 65 basis points to 88.9%, and we now expect to end 2026 closer to 90% occupied. All very positive results from the healthy leasing activity and client demand we are seeing in our markets. We also generated $0.04 per share of outperformance from lower operating expenses in the portfolio. About half of this is from lower repairs and maintenance expense that I anticipate will be deferred to later in 2026 and is embedded in our expense guidance for the back half of the year. The rest came from lower utilities expense related to lighter energy consumption in the Northeast, where we are working hard to fine-tune our buildings to lower consumption and cost every day. We also had lower real estate tax from the receipt of real estate tax abatements this quarter. We continue to aggressively appeal our real estate tax assessments throughout our portfolio and are seeing positive results in certain locations. Looking at the full year 2026, we are raising our guidance for FFO by $0.05 per share at the midpoint, by bringing up the bottom end by $0.09 to $6.99 per share and the top end of our range by $0.01 to $7.05 per share. Strong leasing performance across our portfolio is giving us increased confidence in our growth outlook. In our same-property portfolio, we are increasing our assumption for our share of NOI growth over 2025 by 30 basis points to between 1.8% and 2.6%. The increase mirrors the accelerated occupancy growth that Doug detailed. In our development portfolio, we are increasing our assumptions for NOI by $0.03 per share based on faster lease-up and lower expenses. At 360 Park, as Doug mentioned, we signed 50,000 square feet in the quarter, and we are now in negotiations to lease the last available floor. On the expense side, we started capitalizing expenses at Reservoir Place where we commenced redevelopment this quarter with the signing of our lease with Boston Dynamics. We have been extremely successful in executing our asset sales program, which is raising capital to fund our developments and reduce debt. As Owen described, we are ahead of the expectations we laid out last year at our Investor Day, not in the total volume of asset sales, but in our timing. The accelerated sales timing has a slightly more dilutive impact than the prior guidance that we provided. Including the impact of lower net interest expense from deploying the sales proceeds to reduce debt, we expect the foregone NOI from our sales to reduce FFO by approximately $0.02 per share when compared to our prior assumptions. Lastly, we raised our assumption for fee income revenue by $0.01 per share from higher construction management fee income and leasing commissions earned from our joint venture portfolio. To summarize, we have increased our guidance for 2026 FFO by $0.05 per share at the midpoint to our new range of $6.99 to $7.05 per share. The changes come from increases in our assumption for growth in our share portfolio NOI by $0.06, lower net interest expense of $0.03, and higher fee income of $0.01. These are partially offset by a reduction of NOI from asset sales of $0.05. Overall, we had a great quarter, and all phases of our business strategy are clicking. We raised both our FFO and occupancy guidance driven by consistently strong leasing volumes and excellent progress on leasing our vacant and near-term expiring space. Our occupancy has now increased for three consecutive quarters, and we are executing on our planned asset sales program to both reduce leverage and redeploy capital into higher-yielding new developments. Operator, that completes our formal remarks. Can you open the lines up for questions?
分析師問答
Thank you, sir. As a reminder, to ask a question, you will need to press 11 on your telephone. To withdraw your question, please press 11 again. We ask that you please limit your questions to no more than one, but feel free to go back into the queue. And if time permits, we will be happy to take your follow-up questions at that time. Please stand by while we compile the Q&A roster. I show our first question comes from the line of Nicholas Yulico from Scotiabank. Please go ahead.
So first question, clearly, the occupancy benefit is picking up in the portfolio, which will help for 2027 earnings. Can you talk a little bit more, Mike, about how the asset sales are going to work in terms of the impact on 2027 versus debt repayments? Since some of the income-producing asset sales are more back half weighted, like 7 Times Square or potentially even next year, is there dilution we should be thinking about for 2027? And then also, regarding capital, any update on whether there might be excess sale proceeds to use for stock buybacks?
On the asset sales side, as I mentioned, we are ahead of plan, and Owen mentioned that as well. The dilution in 2026 is a little bit higher than we had originally stated at our investor day. At the beginning of the year, we said dilution would be $0.06 to $0.09. If we get everything done that we expect, it will be closer to $0.11. The majority of our asset sales will be completed; we will evaluate incremental sales going forward as well. Our goal remains to bring down our leverage into the low 7s, which gives us optionality for future investment activities. Those future investments could include new developments or stock buybacks, and we will evaluate all of those options. With respect to 2027, we are not giving guidance on 2027 right now. The total asset sales that we project are still $1.9 billion by 2028. As Owen described, we could have $1.7 billion done potentially by the end of this year, which means that next year will be lighter.
Thank you. I show our next question comes from the line of Steve Sakwa from Evercore ISI.
Yes. Thanks. Good morning. Given the leasing success and the faster ramp in occupancy, how are you thinking about the ultimate stabilized occupancy rate of the portfolio? Has that changed in your mind? Has the timing of that stabilization been pulled forward given current leasing activity?
Steve, right now we are sticking to our target of 91% occupied at the end of 2027. If things continue on the same trajectory, we would be more aggressive, but we are not ready to change guidance yet. Looking at our lease expirations and available space, the concentration of vacancy is in two main areas. The first is Embarcadero Center in San Francisco, where we have the most short-term opportunity to exceed projections, which would likely occur in late 2027 or early 2028. The second is our portfolio available space in tertiary markets, in the urban edge of Boston and in our West Los Angeles portfolio in Santa Monica. The value of that space is less than CBD properties like Midtown Manhattan, the Back Bay of Boston, or San Francisco. My estimate for the long-term maximum occupancy is somewhere between 94% and 95%. By the end of 2027, we think we will be at 91% or maybe a little better, and in 2028 we could approach that higher number. We will always have some marginal availability given the ten-year lease cadence and large clients that may relocate, so I do not expect much above 94%–95%.
Thank you.
Congrats on a great quarter. In the prepared remarks, you touched on some price discovery. Can you walk us through what you're seeing in the transaction market with fundamentals improving but higher interest rates impacting pricing on land, residential, and office?
Transaction volumes for office are off the bottom and have grown significantly over the last year, but they remain well below pre-COVID levels. We are in recovery mode. A large percentage of buying is family office and opportunistic capital seeking discounts to replacement cost; that has been the majority of transactions and that is logical during recovery. The deals I mentioned this quarter mirror last quarter: cap rates around the 7% range with the possibility of stabilizing slightly higher. The transactions I referenced are among the better deals and are not necessarily true premier workplace trades.
I show our next question comes from the line of John Kim from BMO Capital Markets.
Thank you. Owen, you mentioned at Reservoir Place you are expecting a cash return of over 10%. Is that on the incremental CapEx or does it include historical cost of the asset? And going forward, what is your hurdle rate on development, for example, on build-to-suit developments similar to 725 12th Street?
The 10% I mentioned includes an inferred value for the building that was taken out of service. The cash yield on the incremental capital would be materially higher. Our target yields depend on market, preleasing, and risk, but generally we are getting 8%+ yields on our development. For 343 Madison we remain on track for the 7.5%–8% stabilized unleveraged cash return. Our Washington deals generally pencil over 8%, which is accretive to where the stock is trading relative to cap rates.
I show our next question comes from the line of Anthony Paolone from JPMorgan. Please go ahead.
Thanks. Doug, you mentioned an opportunity at Embarcadero Center in the near term. If momentum in Northern California persists, what do you think BXP's biggest opportunities are there? What are you likely to do with that portfolio?
I'm going to let Rodney answer that because he has some interesting opportunities, one of which is physically ours and others we are working on that he can discuss.
The market is very strong in Northern California. We are taking advantage of increased demand from the AI sector. The pipeline of tenants in the market is pushing 9 million square feet, which is unprecedented. There is limited premier workplace supply; tenants looking for 50,000 to 100,000 square feet of top-tier space have very few choices. That is prompting discussions about building new buildings. We have been awarded, through a competitive assignment, a development consultant role on a downtown site we are familiar with from the past cycle and we will have a role in that. We have an opportunity to invest if the market and demand support it. At 4th and Harrison we have a potentially phased project of roughly 800,000 square feet that we were ready to start before COVID. It is adjacent to where many AI companies in Mission Bay are located. If demand holds, we could pursue a project there, not speculative but with user discussions ongoing.
To summarize for Anthony, we are involved in a couple of interesting opportunities in the San Francisco CBD, not the Peninsula. If market rents reach a point where new construction makes economic sense, we have places where we can create new premier product for clients.
I show our next question comes from the line of Michael Goldsmith from UBS. Please go ahead.
Good morning. The recovery story has been occupancy-led, but the message this quarter felt more rent-growth oriented. Is that correct? Can you talk about pricing power—whether it is increasing and if that is across all markets or just the strongest ones?
For us, the occupancy story is primary because occupancy gains translate directly to cash. Mark-to-market rent increases are smaller by comparison. I'll let Hilary discuss pricing power in Manhattan and Bryan discuss our view in Boston, where much of our rate increases will come from.
Pricing power in Manhattan remains quite favorable to landlords and it is expanding geographically. Strength has been strongest in the best submarkets of Midtown and continues to expand outward to other Midtown submarkets and Midtown South. We have now spoken for every single floor at 360 Park Avenue South and are seeing landlords across Midtown South post higher rents as they lease remaining vacancy. In Midtown proper, we are receiving inbound interest at our highest-quality building and at 343 at rents consistently 10%–15% above last year. In lower stacks of our buildings where rents are more affordable, we are seeing 20% increases year over year due to lack of available space. There is great strength from the landlord perspective in New York City.
In Boston, our rent roll snapshot shows we are 97%–98% leased in the Back Bay, 98% in Cambridge, and 100% in Cambridge lab. Compared to the competitive set, there's a significant spread between general Class A vacancy and our competitive set—sometimes as much as nine points. For us, price discovery will occur mostly in renewals because we have little vacant space to take to market. We are educating the marketplace, brokerage community, and clients about factual comps, and we anticipate pricing power in Boston.
Our next question comes from the line of Seth Bergey from Citi.
This is Nick Joseph on for Seth. On mark-to-market for the West Coast portfolio, how do you think about where the portfolio sits today versus market rents?
It is building-specific; I'll use San Francisco as a poster child. In Mountain View, we had a significant markdown because previously we had been getting very high rents and now new leases are closer to $4–$5 per square foot per month, still high but lower than the peak. At Embarcadero Center, the situation is neutral to positive. Upper floors and newer product have embedded market opportunity for growth. Lower portions of EC 1–3 with older second-generation space are more competitive and may see slight markdowns. At 680 Folsom, 535 Mission, and Salesforce Tower, we expect material increases in mark-to-market. For example, at 680 Folsom our asking rents are above the rents set to expire in 2028–2029. Salesforce Tower has embedded growth of roughly 30%–40% and we have about 200,000–250,000 square feet of expirations in 2027–2028, presenting real opportunity. Seattle and West LA are more muted: Seattle modestly lower, West LA still struggling. Those markets are small portions of our portfolio, so they are not material to overall company performance.
I show our next question comes from the line of Blaine Heck from Wells Fargo. Please go ahead.
With respect to 343 Madison, can you elaborate on the appetite from potential equity partners, the timing we should expect on those sales of interest, and any color on how you and those partners are thinking about value versus expected cost on the 30%–50% interest you plan to monetize?
We have a letter of intent with an investor to purchase a 10% interest in the project and expect it to close this quarter. We continue to talk to additional investors about selling additional interest up to 30%–50%. We are selling down interest in what we consider to be one of the best office developments in the U.S., and are seeking favorable pricing and governance terms. In high-level terms, our developer yield on the project is around 8% and we expect the delivered property's value to reflect a cap rate in the 5.5%–6% range. As we monetize interests, we will be moving gradually from an 8% yield towards that 5.5%–6% yield.
Next question comes from the line of Caitlin Burrows from Goldman Sachs. Please go ahead.
Earlier you mentioned that 48% of leasing in Q2 was renewals, extensions, and expansions. Can you talk more about renewal rates—what retention has been over the past three years and if it's fair to expect it to increase going forward?
This is a timing-dependent answer. As we get closer to a lease expiration, retention rates can appear to come down because larger transactions have already occurred earlier. Historically, our retention has been in the 45%–50% range. That is largely because we are often fully leased and cannot always accommodate client growth. Right now, our near-term expirations look favorable: for 2027 expirations we have good clarity on much of the activity, and I would not be surprised if we get above that 50% level for the portfolio in front of us. But on a long-term basis, 45%–50% is a reasonable expectation.
To add, in the last couple of quarters we've had several larger lease renewals that we signed a year or two ago commence, and our commenced leases over the last two quarters have shown retention closer to 60%–65% for those cohorts. Long-term, though, it averages around 50%. This year's better retention is reflected in the occupancy growth we're seeing.
Our next question comes from the line of Floris Van Dijkum from Ladenburg Thalmann. Please go ahead.
Thanks. Appreciated the visibility on the signed but not occupied pipeline. Can you quantify what that signed occupancy growth would be in terms of NOI? New York signed-but-not-occupied is different than Los Angeles or D.C., so some color on the NOI impact by market would be helpful.
I don't have the exact breakdown in front of me, but the majority of the signed-but-not-occupied space in 2026 is in Manhattan—largely coming from 360 Park Avenue and 200 Fifth Avenue. That is where the most of the near-term NOI impact will come.
Next question comes from the line of Upal Raina from KeyBanc Capital Markets. Please go ahead.
Doug, you said Embarcadero Center could give the most short-term uplift in occupancy. Can you update us on the pipeline for those buildings and any timing you can share?
I'll give a brief overview and let Rodney add color. Big picture, the market there is granular with many smaller financial services and professional services users, which leads to many partial-floor transactions. Filling available space takes time, but we are seeing activity.
One of our key strategies is building prebuilt space. For example, we have two floors at 1 Embarcadero Center under construction now—one to cater to tech build-outs and another for law and professional services. We already have interest on both and expect success by investing ahead to get spaces ready for occupancy. Shell-condition second-generation spaces will be the hardest to lease, so proactive investment is how we'll win. Embarcadero Center will also benefit from the Embarcadero Plaza park build, a private-public partnership with the City of San Francisco that will enhance the environment around the Center.
Next question comes from Dylan from Green Street.
Thanks. Any possibility the ultimate asset sales goal ends up being much higher than $1.9 billion? And once you're done with the program, does that leave the portfolio largely trophy and Class A, or would there still be 5%–10% noncore remaining?
We will continue to sell assets as appropriate. The cadence will likely slow over time. Three categories to consider: land—additional residential entitlements will take time to monetize and will continue beyond 2026; built and close-to-stabilized apartment buildings—we still have a couple that we have not sold; and a handful of office assets that are not yet stabilized or are in lease-up. As those properties are leased up and we can maximize value, we will dispose of them. So dispositions will continue but at a slower cadence beyond 2026.
The 'land' bucket includes older suburban office buildings where we have decided the recovery will lag the opportunity for residential entitlement. Many jurisdictions are favorable to housing now, and we have over a million square feet of suburban space that will ultimately disappear from our portfolio. The monetization of these will generate hundreds of millions of dollars over time, not small amounts.
Next question comes from the line of Richard Anderson from Cantor Fitzgerald.
AI has been a demand driver, but it reminds me of the life science boom several years ago that did not turn out great in all cases. Are there lessons learned from that experience and how are you approaching AI demand today with any guardrails?
The future of AI and its impact are difficult to project. The primary leasing benefits we are seeing are not only from AI companies directly but from market tightening—AI firms are creating demand that displaces other tenants and drives upgrading and relocation activity. Our core service clients who support AI are also growing. If AI were to decline, that would be negative, but much of the leasing benefit is indirect. When we do lease to AI companies, we focus on credit quality and obtain letters of credit where appropriate and monitor exposure to early-stage AI startups.
To differentiate from the life science cycle: a lot of speculative lab buildings were built with costly lab infrastructure on spec, which created a supply issue. AI demand is for office space—BXP is not a data center developer—so it's a different dynamic. We're not seeing speculative CBD office construction in our markets. The life science overbuilding in certain suburban lab markets was fundamentally different, which is why today's market dynamics differ.
Next question comes from Dylan Abramowitz from Deutsche Bank.
Mike, last quarter you talked about leasing CapEx around or above $400 million for the year and you were on pace from first-half spending. Could you update where that number might shake out for 2026 and views for 2027? How does leasing CapEx trajectory impact AFFO growth in the back half and beyond?
We continue to sign additional leases and are increasing occupancy projections for 2026. That leads to more leasing transaction costs this year. I expect leasing CapEx will be closer to $500 million than $400 million based on current activity. That will impact AFFO in 2026 since many leases have free rent periods at inception, which depresses cash in the near term. However, those leases will become cash-paying in 2027 and will positively impact AFFO going forward. So 2026 will see higher transaction costs and elevated straight-line rent, and 2027 will show the cash flow benefits.
Next question comes from the line of Alexander Goldfarb from Piper Sandler.
As the company strategizes for 2027, is the priority keeping earnings growth accelerating first and debt payoff second, or is debt reduction prioritized, given your strong position and the tailwinds?
Our clear priority is growing FFO per share. We will continue to sell assets where appropriate when we think we can achieve fair value and optimize the portfolio. A large portion of sales have been land and apartments which trade at accretive yields, mitigating dilution. We recognize the importance of earnings growth and are balancing capital allocation between deleveraging and redeploying into higher-yielding developments.
I agree with Owen. That is our goal.
Next question comes from the line of Brendan Lynch from Barclays. Please go ahead.
Are there other buildings in the portfolio like Reservoir Place that could capture demand for similar full-building redevelopments, and how do redevelopment yields compare to other uses of capital?
Yes, there are other opportunities. These require time, effort, and local diligence. Our Boston team has done this twice, including Analog Devices at 105 Winter Street and now Reservoir Place. We have buildings in Northern Virginia that could produce similar outcomes. These are not imminent necessarily but are physical opportunities. We are also actively matching clients with buildings to create opportunities outside our portfolio.
Good morning. As Doug and Owen alluded to, we are working on what we hope will be the third opportunity here in Downtown D.C. with inbound clients. The key has been matching client size with building size to make highly leased developments from the outset. There are many sites and law firms interested in similar projects, and we are having conversations across clients, brokers, site owners, and lenders. The group of players that can execute on these transactions is relatively small, and we are fielding interest.
In Boston, suburban activity is an additional factor. We have captured about 70% of leasing in the Waltham market over the last year and a half. Reservoir Place is a great building with premier attributes and a location that is hard to replicate. Combined with our ability to deliver bespoke design and timing for clients whose uses are different than conventional office, we've had a competitive advantage. That capability mirrors what Peter and others see in D.C.
Next question comes from the line of Ronald Kamdem from Morgan Stanley.
On same-store NOI, the cash number was reiterated at roughly flat for the year. What were the drags this year and how should we think about cash same-store ramp as you flip the calendar with occupancy tailwinds?
Cash same-store will lag GAAP same-store as we gain occupancy because many of these leases include free rent at inception. We increased GAAP same-store guidance by 30 basis points due to higher occupancy but did not move cash because of free rent periods, generally ranging from 6 to 12 months. Expect the cash benefits to materialize in 2027 when those leases begin paying cash rent, and that's when cash same-store will catch up with GAAP same-store.
Our last question comes from the line of Vikram Malhotra from Mizuho. Please go ahead.
With attractive debt markets, would you consider taking any unencumbered assets to market utilizing current conditions? And on capital allocation, given the San Francisco turn, is there an opportunity to deploy more capital now, similar to prior JVs for value-add office?
Secured and unsecured debt markets are both attractive, and the bank market is active. A high-quality CMBS execution or an unsecured bond could price at tight spreads today. We are focused on refinancing debt as it comes due and evaluating the best available options across unsecured bonds, CMBS, bank loans, and even convertible debt. If we were to issue incremental debt, we would weigh secured versus unsecured opportunities and choose the best approach based on needs. Regarding acquisitions, we evaluate all opportunities. The bar is high: any acquisition must allow us to convert an older building into a premier workplace with returns that meet our development yield targets, which are generally around 8% or higher. If we can find such opportunities, we will pursue them.
To add, the bar for buying an older building is that we must believe we can convert it into a premier workplace at an attractive yield relative to development. If we find such opportunities, we will act.
That concludes our Q&A session. At this time, I would like to turn the conference back over to Owen Thomas, Chairman and Chief Executive Officer, for closing remarks.
Well, it has been an hour and 22 minutes, so we have nothing else to report. Thank you all for your interest in BXP.
This concludes today's conference call. Thank you for participating. You may now disconnect. Good day.