Avalonbay Communities Inc (AVB) — Q3 2025 Earnings Call Transcript

Thank you, Bahn, and welcome to AvalonBay Communities Third Quarter 2025 Earnings Conference Call. Before we begin, please note that forward-looking statements may be made during this discussion. There are a variety of risks and uncertainties associated with forward-looking statements, and actual results may differ materially. There is a discussion of these risks and uncertainties in yesterday afternoon's press release as well as in the company's Form 10-K and Form 10-Q filed with the SEC. As usual, the press release does include an attachment with definitions and reconciliations of non-GAAP financial measures and other terms, which may be used in today's discussion. The attachment is also available on our website at investors.avalonbay.com, and we encourage you to refer to this information during the review of our operating results and financial performance. And with that, I will turn the call over to Ben Schall, CEO and President of AvalonBay Communities, for his remarks. Ben?

Thank you, Matt. I'm joined today by Kevin O'Shea, our Chief Financial Officer; Sean Breslin, our Chief Operating Officer; and Matt Birenbaum, our Chief Investment Officer. Before discussing our Q3 results, which were below our prior expectations and our updated outlook for 2025, I want to start by emphasizing a series of AvalonBay tailwinds and strengths that keep us confident in our ability to drive superior earnings and value for shareholders. First, our portfolio with its heavy concentration of communities in suburban coastal markets continues to be well positioned. With a more uncertain demand backdrop, we believe that those markets and submarkets with lower levels of new supply will continue to be the relative winners. Our established regions are particularly well situated with deliveries as a percentage of stock projected at only 80 basis points next year. And given how challenging it is to get new development approvals and the amount of time it takes to get those approvals, we expect our markets to continue to benefit from below-average levels of supply for a number of years. A second differentiator for us is the $3 billion of projects we currently have under construction, which will generate a meaningful uplift in earnings and value creation in 2026 and 2027. These projects are tracking ahead of our initial underwriting and, importantly, are benefiting from reduced construction costs, which translates into a lower long-term basis for shareholders. These projects are 95% match funded with capital that we previously raised through a mix of equity and unsecured debt with an initial cost of capital of below 5%, providing an attractive spread to our development yields on these projects. Third, our balance sheet is in terrific shape with low leverage and over $3 billion of available liquidity. As we look ahead, this balance sheet strength provides us with the flexibility to continue to redeploy free cash flow, disposition proceeds and low-cost debt into our next set of accretive development projects, as well as to buy back our stock when appropriate, as we did in Q3, having repurchased $150 million of our stock at an average price of $193 per share. Finally, we continue to advance on our set of strategic focus areas, which are generating incremental earnings and cash flow from our existing portfolio as well as on new developments and acquisitions. This year, we've made strong progress in advancing toward our longer-term portfolio allocation targets with a continual eye towards enhancing the cash flow growth of our portfolio. And we remain very excited about our progress on our operating model initiatives, including the expanded set of uses for technology, AI and centralized services. By year-end 2025, we expect to be roughly 60% of our way toward our target of generating $80 million of annual incremental NOI from these operating initiatives. Turning to the third quarter. Slide 5 in our earnings presentation summarizes our Q3 and year-to-date results. We are on track to start $1.7 billion of development projects this year with a projected yield in the low 6s on an untrended basis. We've also completed our planned capital sourcing activity for the year, having raised $2 billion of capital at an average initial cost of 5%, generating a spread north of 100 basis points relative to development yields. Slide 6 provides the breakdown of third quarter core FFO relative to our prior expectations. Of the $0.05 underperformance relative to our outlook, $0.03 was attributable to same-store portfolio results, of which $0.01 related to lower revenue and $0.02 related to higher operating expenses, including in repairs and maintenance, utilities, insurance and benefits. Turning to Slide 7. Apartment demand has been softer than anticipated this year, which we attribute mainly to the reduced job growth backdrop with related factors, including higher macroeconomic uncertainty, lower consumer confidence and a reduction in government hiring and funding. As shown on the left side of Slide 6, the National Association of Business Economics, or NABE, is now projecting growth of 725,000 jobs in 2025, down from the over 1 million jobs in their prior forecast. And for Q4, NABE is projecting growth of just 29,000 jobs per month. We revised our revenue expectations as part of our midyear forecast with results for July and August generally tracking to those expectations, as shown on the right-hand side of Slide 7. As Sean will discuss further, softness on rental rates in August continued in September, along with a slight occupancy dip. With further softness continuing into October, trends that are now incorporated into our updated outlook for the remainder of the year. As shown on Slide 8, we've also updated our expense outlook for the year to 3.8%. After benefiting from meaningful operating expense savings in the first half of 2025, we've had trends run against us across a set of expense categories without any offsetting savings. For example, in repairs and maintenance, we knew that certain savings from the first half of the year would be incurred in the second half, but have incurred more and higher cost repairs and non-repeat projects than anticipated. Other unfavorable variances include insurance, utilities and associate benefit costs. Given our Q3 results and these revenue and operating expense trends, we've updated our outlook for the full year, which Kevin will now discuss in more detail.

Thanks, Ben. Turning to Slide 9. We present our updated operating and financial outlook for full year 2025 as compared to our prior outlook on our second quarter call and on our initial outlook for the year that we provided in February. We are lowering our full year core FFO per share guidance by $0.14 to $11.25 per share, which reflects an updated expectation for year-over-year earnings growth of 2.2%. Our updated full year outlook reflects same-store residential revenue growth of 2.5%, same-store residential operating expense growth of 3.8%, and same-store residential NOI growth of 2%. Turning to Slide 10. We highlight the components of our updated outlook for full year core FFO per share in the second half of the year for key parts of our business as compared to our prior outlook on our second quarter earnings call. Specifically, as Ben previously mentioned and as detailed on this slide, our third quarter core FFO per share results were $0.05 below our prior outlook. And for our fourth quarter core FFO per share, we provide a comparison between our prior outlook and our current outlook. The expected $0.09 decrease is primarily driven by $0.06 of lower NOI from the same-store portfolio, consisting of a $0.04 decrease in same-store residential revenue and a $0.02 increase in same-store residential operating expenses. The remaining $0.03 reflect lower expected earnings contributions from lease-up NOI, commercial NOI, joint ventures, and other stabilized NOI. Taken together, our 3Q results and revised fourth quarter outlook resulted in an updated outlook for the full year core FFO of $11.25 per share. And with that overview of our updated outlook, I'll turn it over to Sean to discuss operations.

All right, thanks, Kevin. Moving to Slide 11. As Ben noted, we started to experience some softening in key revenue drivers during the quarter. In Chart 1, economic occupancy was generally consistent with our expectations in July and August, but fell below our previous outlook in September and has continued to be below our previous expectation for October. Similarly, rent change started to trend below our midyear outlook in August, driven primarily by weaker move-in rents, which are depicted in Chart 3. While move-in rents softened across most regions, the deceleration was most pronounced in the Mid-Atlantic and Southern California, which was driven by L.A. and Denver. In terms of underlying bad debt, while we ended the quarter close to our original estimate, we experienced an uptick in August, which contributed to the unfavorable revenue variance for the quarter. Turning to Slide 12. We now expect same-store revenue growth of 2.5% for the full year 2025, down 30 basis points from our midyear outlook. The primary drivers of the reduction are average lease rate, which is estimated to account for 20 basis points, along with economic occupancy and underlying bad debt, which are projected to be about 5 basis points each. Our established regions are projected to produce 2.7% revenue growth, while the expansion regions are forecast to be modestly positive. As I mentioned on the previous slide, while the softness we've experienced has been somewhat broad-based, it has been most pronounced in the Mid-Atlantic and L.A. The Mid-Atlantic has been choppy since the second quarter, and it softened further during Q3 as the probability of a government shutdown increased. Given the shutdown has become a reality and is heading into a second month in a couple of days, we expect continued weakness in the region through year-end. And in L.A., job growth in the film and television industry has continued to be weak. It's been estimated that the number of film and television jobs in L.A. has declined by roughly 35% compared to just 3 years ago. And stage occupancy, which reflects the percentage of time sound stages are being used by production companies in the region, has been trending in the mid-60% range recently, down from 90%-plus levels just a few years ago. While new tax incentives were passed in July this year to support film and television production in California, any employment benefit from them won't likely be realized until sometime in 2026 or beyond. Moving to Slide 13. As we start thinking ahead to 2026, while job growth has been below expectations recently, our portfolio is positioned to perform relatively well given 2 important factors: First, the very low level of new supply expected in our regions; and second, a lack of affordable for-sale alternatives. New supply in our established regions is expected to decline to roughly 80 basis points of existing stock in 2026, which is not only less than half the trailing 10-year average but also a level we haven't experienced since 2012. It's also roughly consistent with what occurred during the '90s decade, which was a terrific time period for us. On the right side of Slide 13, although mortgage rates have been trending down recently and home values have flattened out or declined in many regions, for-sale housing remains very unaffordable in our established regions. It still costs almost $2,500 per month more to own the median-priced home relative to the median apartment rent in these markets. Overall, while we don't have a crystal ball regarding job and wage growth for 2026, again, our portfolio is relatively well positioned for any demand environment given the supply picture and the lack of affordable alternatives. And as it relates to our portfolio and the setup for 2026 revenue growth, we're currently projecting our earn-in to be roughly 70 basis points. Additionally, we continue to expect improvement in underlying bad debt as we work through the backlog of cases in several established regions and our various screening tools further constrain new entrants to the bad debt pool. Forecasted benefit for the calendar year 2025 is approximately 15 basis points. For 2026, I would expect at least 15 basis points and likely more given some of the underlying activity we're seeing across the portfolio. And third, while it won't likely be as strong as the last couple of years as we begin to stabilize our AvalonConnect offering for residents, we still expect another well above average year of growth in other rental revenue in 2026. Now I'll turn it to Matt to address our development activity.

Thanks, Sean. Turning to our development activity. As shown on Slide 14, our current lease-ups continue to perform better than our initial expectations, reflecting the conservative underwriting approach we take, where we do not trend rents and analyze every new start primarily on its current economics. Our development underway reached $3.2 billion by the end of the third quarter, was 95% match funded and underwritten to an untrended yield on cost of 6.2%. We opened several new lease-ups over the summer and now have 6 communities where there is enough leasing activity for us to update the rents and yields to current market. This $950 million in development activity is running 10 basis points above the initial projections, thanks to $10 million in cost savings and rents that are $50 per month higher than pro forma, generating a further lift to the value creation and earnings accretion these communities will deliver once they are complete and stabilized. We have another 3 communities, all in New Jersey, that are just starting their lease-ups and rents at those assets are currently set at 2% above pro forma. So the trend of development outperformance is likely to continue as all 9 of these communities look to complete their lease-ups next year. And the 13 communities that won't start lease-up until 2026 or '27 should open at a time when there will be much less competitive new supply, as Sean detailed earlier. Turning to Slide 15. We are strategically increasing our development underway when the industry as a whole is retrenching, taking advantage of the benefits of our integrated platform to build at a time when costs are lower and competition is more subdued. As we look to 2026, many of these favorable tailwinds should persist, although we are also mindful of the softening revenue environment and the associated impact on our cost of capital to fund new starts going forward. And with that, I'll turn it back to the operator for Q&A.

Maybe following up on Matt's development comments. Just curious kind of how you're looking at the next crop of projects and properties, kind of how you're thinking about those? And maybe also comparing that with you guys were active on share repurchases in the quarter. If you could kind of talk to those capital allocation decisions.

Sure, Jana. Thanks for the question. I want to emphasize the strength of our balance sheet, which is in excellent shape and provides us with significant flexibility in our capital allocation decisions going forward. We currently have a robust array of reinvestment opportunities within our existing portfolio. This year, we are actively making revenue-enhancing investments and anticipate similar opportunities for the coming year. Regarding development, we are currently considering development starts in 2026 around $1 billion, based on our pipeline and available opportunities. These projects are mostly located in our established regions where operational fundamentals remain stable. We're also experiencing strong construction buyout savings in these markets. Based on current rents and costs, while these projects aren't starting right away, the expected yields on that $1 billion are in the range of 6.5% to high 6%, providing a solid spread for raising additional capital. We will continue to be flexible and adjust as necessary since we approve each development project individually. We have indeed raised the target returns we expect from developers for next year, but we remain optimistic about having another productive year in development activities. Additionally, our strong balance sheet gives us the option to buy back stock as we did in the third quarter, which allows us to invest effectively into our existing portfolio. This is the current landscape as we consider our capital allocation decisions.

I appreciate all the comments on the markets. Maybe for Ben. Just as you talked about SoCal and then obviously, the government shutdown won't go forever, but I mean, do you kind of look at those markets maybe differently just on a long-term basis? And would your, I guess, preference to have lower exposure in both of those markets kind of on a go-forward basis?

Yes. I'll start with an overview and then let Sean discuss the situation on the ground, Steve. Overall, we are making progress on our portfolio allocation targets, which includes reducing our presence in the Mid-Atlantic and California. Additionally, we have set targets not just at the market level but also within our regions. For instance, following our recent sales in D.C., we are focusing on increasing our exposure in the Mid-Atlantic, particularly in Northern Virginia for various reasons. After that transaction, nearly 50% of our portfolio is now in Northern Virginia. This approach reflects our strategy of balancing long-term planning with short-term transactional activities. Sean, would you like to share your observations in both markets?

Yes, Steve. The government shutdown was a concern in the third quarter, and it has now become a reality. We've encountered similar situations in the past, and this typically reflects cyclical trends rather than a permanent change. As Ben mentioned, we're pleased that nearly half of our portfolio is currently in Northern Virginia, which is performing better than the district and some Maryland markets. It's important to note that we expect to deliver around 15,000 new units in the Mid-Atlantic in 2025, which is anticipated to drop to just 5,000 units across the entire DMV in 2026. If we move past the current situation into a more stable, even modestly growing job environment, this could set up a favorable scenario for revenue growth. In Southern California, while the economy has experienced cycles, it remains diversified. The entertainment sector has faced challenges, but certain tax incentives and activities may help rejuvenate it in the future. This region usually sees relatively low levels of new supply, which we expect to continue in the next few years, despite being a significant market. Thus, we're considering both short-term decisions and long-term strategies, mainly focusing on shifts within the region rather than outside of it.

It's Nick Joseph on for Eric. Maybe just going back to the capital allocation answer earlier. You mentioned development starts maybe in the mid-6s to high 6s. I think buybacks would be somewhere around the mid-6s now. How does that compare to what you're seeing kind of real time in the transaction market? Have kind of going-in yields changed at all given some of the weaker rent growth assumptions? And as you think about that, is there also a difference within some of the different markets that you're looking at today?

It's Matt. I'll take that one. The short answer is we haven't seen any change in how the market is pricing stabilized asset sales. It remains in the mid- to high 4% cap rate range to low to mid-5% cap rate range depending on the location. Our own activity reflects this as well. For example, D.C. is on the higher end of that range, while suburban Seattle is on the lower end, where we recently sold an asset this quarter at a 4.6% cap rate. It's been pretty stable. Additionally, as long rates have slightly decreased, that has boosted buyer confidence. Transaction activity in multifamily is significantly higher in Q3 compared to Q3 of '24. However, the assets receiving bids are mainly those with decent rental momentum or at least not declining. So, cap rates and values are holding firm so far.

Matt, I just wanted to follow up quickly. Did I understand your comments about the D.C. cap rates correctly, that the properties sold before dispositions were around a low-5 cap? And then Kevin, can you explain what caused the surprise in repair and maintenance? Are we facing labor availability issues or was there something else that went wrong in the repair and maintenance functions?

Yes. John, it's Matt. The cap rate on the DC sales was probably mid-5s. There was a little bit of retail in the portfolio. So the residential cap rate was probably kind of low to mid-5s, but I'd say the overall transaction was right around 5.5%, and then...

John, it's Sean. On the R&M side specifically, it's kind of a smattering of different things. What I'd say at a high level is we had a pretty good experience going through Q2, as Ben mentioned in his opening remarks in terms of the benefit, we expect a little bit of that to come back. But unfortunately, just kind of hit a bad streak in Q3 in terms of various accounts that came through on the repairs and maintenance side, slightly higher cost per turn in terms of the way the units came to us. Some of them are skips and a VIX, the higher cost. So I wouldn't say there's one particular pattern there other than we just probably underestimated a little bit kind of where we land in Q3 relative to what happened in Q2.

I think last quarter, there was a discussion around lease-up at an asset or 2 in Denver development assets. Just wondering if there was any update there for those assets. And then I guess just more broadly, if you sort of think about the performance of development assets and lease-up, how are they doing? And as sort of some of them maybe from a year ago or 9 months ago, as you get closer to sort of that annualizing the initial leases, what is sort of the performance in terms of people at renewal given sort of the pace of lease-up?

Adam, it's Matt. I'll begin by addressing that, and then Sean can provide further details on our Denver lease-ups. Overall, our lease-ups are performing well, and we are seeing some outperformance in rent. Importantly, we are also achieving significant cost savings, which are sustainable over time. While the net operating income will fluctuate, the reduced basis is permanent. Just considering the deal we finalized in the first half of this year and Annapolis, which we completed this quarter, along with another project in Maryland next quarter, we are looking at approximately $12 million in cost savings over 1,000 units, equating to $12,000 per unit in lower basis. This is quite compelling. I anticipate that our yield outperformance going forward will come from both the denominator and the numerator, although we are still seeing some benefits in the numerator. Typically, we lease at a pace that allows us to fill the assets within 12 months to avoid competing with ourselves on renewals, and we have generally succeeded in this strategy. There have been a few exceptions, particularly in Denver, where the Governor's Park submarket south of downtown is overwhelmed with supply. So Sean, would you like to share more on that and compare it to other areas?

Yes. Just to give you some insights on Denver, we really had 2 lease-ups. One just finished and stabilized at the end of the third quarter, which is in Westminster. And then the other one is Governor's Park, which Matt has referenced. So, on average, they did about 20 leases a month during the third quarter, which is a little bit below where we typically would like. But again, Westminster was heading right to stabilized mode. So a little bit softer pace there. And then concessions on the Westminster deal averaged about 150% of a month's rent, and it was more than 2 months rent at the Governor's Park deal. So certainly a soft environment in Denver. I think that's pretty apparent to pretty much everybody nowadays. But fortunately, we have one that's stabilized and the Governor's Park deal is approaching 90% leased at this point. So we're getting pretty close.

The other piece of good news I would add, and I believe I mentioned this last quarter as well, is that a significant portion of our lease-up activity now and looking into next year is in the suburban Northeast, which remains quite strong. As I noted, we have three lease-ups currently leasing in New Jersey and a fourth that is nearing the end of its lease-up, and that market has been very solid.

Just thinking through potential share repurchase activity from here, I guess, do you have capital lined up to fund that today? Or would you need to source additional capital to fund future purchases? And just wondering if dispositions are still the best avenue for that today?

Sure. Austin, this is Kevin. I want to highlight a few key points for you. First, our balance sheet is in excellent condition. As shown in our earnings release, our leverage stands at 4.5 times net debt to EBITDA. If we account for the nearly $900 million in forward equity, our leverage effectively drops to around 4 times. We have almost full access to our line of credit, with only about $200 million in commercial paper. This means we have ample liquidity and leverage capacity if we choose to utilize it. Regarding our share repurchase program, we have reauthorized it, providing an additional $500 million in authority for future buybacks. As Ben mentioned, we're ready to be flexible. We are currently planning for approximately $1 billion in starts next year. However, the choice between buybacks and development isn't strictly one or the other. As evidenced by our actions in the third quarter, we continued to invest in development while also repurchasing $150 million in shares. While I can't confirm our exact actions for the future today, we have the option to pursue a buyback if it makes sense for us. We acknowledge the attractiveness of our shares but also see the value in continuing development, which provides us with new assets that have a low capital expenditure profile and strong growth potential. There are compelling reasons to keep advancing our development projects, but we also have the capacity to execute a buyback when appropriate. In terms of long-term funding, we would likely utilize our available liquidity through commercial paper, which is currently priced in the low 4% range. Ultimately, our funding decisions will be influenced by our gains capacity, which typically allows for around $500 million in asset sales per year, and we intend to finance any buybacks on a leverage-neutral basis while considering incremental long-term debt. We believe we have enough flexibility to act constructively, and we're prepared to respond to market signals as needed.

As we've been listening to the calls throughout the day, the #1 incoming question is just how do these residential companies have visibility on where the market is going, first for guidance through year-end, but also just to kind of get through the spring leasing next year. So I know there's only 2 months left in the year, so that's probably an easier part of the question. But just as you think about your crystal ball setting guidance and even thinking about where this cycle could go before it gets better, can you point to some of the things that are giving you confidence or that people should be thinking about or that you're thinking about and watching because every company is certainly taking numbers down or their outlook is down given September and October.

I can start by highlighting a few key aspects. First, our portfolio positioning is strong, especially considering the currently low levels of supply, which we expect to persist into next year. As Sean mentioned, it's worth reiterating that we don't require a significant increase in demand due to this supply landscape to achieve strong results as we move into next year. Regarding the overall job market, we hope to enter a phase of greater certainty on the macroeconomic front, including more clarity around tariffs and the conclusion of the government shutdown. This increased certainty should foster greater confidence. Additionally, the current rate dynamics may encourage further investment, allowing businesses to invest more in their operations and workforces as we assess the job market.

I wanted to discuss bad debt, which is slightly higher than you anticipated. I'm curious about the potential reasons for this and whether a significant portion of the bad debt originated from recent development lease-ups.

Yes, John, it's Sean. Yes, the miss in bad debt we referenced is in the same-store pool. So the development assets wouldn't be in that book. And it's really a relatively modest number. It's about 5 basis points different in terms of what accounted for the variance. And in a book like that, when you're dealing with court times and dockets and when the share is going to show up and all that kind of good stuff, 5 basis points is a pretty tight margin of error. But obviously, it was a negative variance. So overall, we feel good about where we're headed with bad debt. I can tell you that as you look at where we are now in terms of the number of accounts that we need to work our way through compared to the end of 2024, we're down 20%, 25%. So it's moving in the right direction. It's just a matter of kind of processing people through. So I would expect, as I mentioned in my prepared remarks, as we look forward to 2026, if we're getting about a 15 basis point benefit this year, I would expect at least, if not likely, more than that benefit in 2026 just based on what we're seeing as different cases work their way through the system and our screening tools get more and more sophisticated in terms of limiting the number of new entrants to the pool.

But in general, I know it's not part of the figure with the same store. Do you tend to get higher bad debt on lease-up communities?

Not necessarily. No. If it is, it can be an outlier community here or there. It's really kind of market specific in terms of the type of customers you're dealing with and tools you use. But in general, it's not necessarily an outlier across the development book as compared to the same-store pool.

But just to follow up on the jobs discussion. I guess, as you're having conversations with tenants in the D.C. market specifically, I guess, what are the chances that there's another shoe to drop with respect to sort of delayed impacts from DOGE. And if someone's laid off, there's usually sort of a lag before they think about vacating the unit or stop paying rent or things like that. And then secondarily, there's been a lot of headlines around weakness in the entry-level job market specifically. And that's not a D.C. comment, that's broad-based. But how does that affect your portfolio more broadly? So I guess kind of a 2-parter.

Yes, it's Sean. I'll begin, and others can weigh in as needed. Regarding the D.C. region, I would say any impact related to DOGE and earlier activities this year is likely being felt now due to normal severance and notice periods. Some individuals may have left 3 to 5 months ago, but looking ahead to 2026, the supply situation appears significantly improved, with a reduction down to about 5,000 units, a figure we haven’t seen in D.C. for a long time. On the demand side, if we can move past the shutdown and return to normal operations, we should be in a better position. The key question is what will happen with the furloughs—whether they become permanent reductions is something we haven't confirmed, but it's a possibility. We need more clarity following the shutdown to understand the potential impact, and if there is one, we might see the effects develop over a period of 6 to 9 months. In response to your second question about AI, we feel positive about our overall standing. The average age of our residents is in the mid-30s, which differs from the younger demographics recently highlighted in the labor market. In the coastal markets we operate in, there are many high value-add jobs. For example, in San Francisco and Seattle, the demand driven by individuals with the skills to advance AI remains strong. Recently, I visited both cities, and discussions with our onsite teams revealed that many prospective tenants seeking apartments are from the AI sector, bringing in high levels of education and experience. We remain optimistic about the nature of jobs in these areas continuing to be advantageous compared to lower value-add positions that may decline in service-oriented roles.

Two questions. The first is on the asset sales, the $585 million in the quarter, I think all of those were developments and it was a slight economic loss. Just curious, you guys speak about the value creation. So the economic loss definitely jumps out. So is there anything specific, was it 1 or 2 of these projects that drove that? Or in aggregate, just want to better understand why there was a loss on the sale?

Yes. Alex, it's Matt. Out of the six sales, two were acquisitions from Archstone assets, while four were developed by us, creating a mix. The primary drivers for the economic loss came from two specific communities. One was Brooklyn Bay, an asset we developed in a small and unique submarket in South Brooklyn, which, with an investment of under $100 million, turned out to be less favorable than we had hoped. The other was an asset in NoMa that we acquired from Archstone, and NoMa has consistently posed challenges since our purchase. We acquired over 60 assets from Archstone, most of which have performed well and contributed positively to our overall investment. However, with a portfolio of that size, occasional underperformance is expected. Even considering the $800 million in total disposals this year, including the $600 million mentioned, the unlevered IRR for these 2025 disposals remains in the mid-8s, which reflects decent investment returns. Typically, our average returns are in the low double digits, so this still represents a solid investment performance. As you know, our long-standing track record stands out in the REIT sector, particularly in the multifamily segment over the years.

Yes. Alex, the other element I'd just add on to Matt's commentary is these are a set of assets that have been on our target list for a while now. And we're waiting for asset values to recover to a certain degree to be able to execute. And so in any portfolio, you're going to have some low performers, but we really think about this as pruning those assets out, redeploying that capital into higher growth opportunities.

Okay. And then yes.

Yes. Alex, the other element I'd just add on to Matt's commentary is these are a set of assets that have been on our target list for a while now. And we're waiting for asset values to recover to a certain degree to be able to execute. And so in any portfolio, you're going to have some low performers, but we really think about this as pruning those assets out, redeploying that capital into higher growth opportunities.

It seems like in the REIT sector, people are becoming more discerning about the markets. For example, there is a preference for very specific areas like Westside L.A., the east side of Seattle, or Northern Virginia. As you evaluate your development pipeline and land options, have you made significant adjustments by eliminating sites that no longer align with your desired submarkets? I'm trying to understand how your strategies have evolved compared to your land bank or land options from a few years ago as you prepare for your next round of projects.

Yes, Alex, I would say we've been quite aware of that for the last several years. If you look at our development rights portfolio today, it's almost entirely focused on both bottom-up and top-down approaches. We're seeking the best risk-adjusted returns and aiming to create value with every deal, but we're also focused on developing assets we believe will perform well in the long term. We incentivize this by setting higher target yields for projects in submarkets we consider weaker. We have been moving in this direction for some time. For instance, Ben mentioned our focus in the Mid-Atlantic, particularly Northern Virginia. In Southern California, we are concentrating on San Diego, which is almost an expansion area for us. This quarter, we initiated a significant project in San Diego, and we currently have two deals under construction there along with two more development rights. We haven't started a development project in L.A. County for 5 or 6 years. Our focus is entirely on San Diego, Orange County, and Ventura in Southern California, for example. Similarly, in Seattle, our portfolio is heavily located on the east side, and our development focus has been exclusively on that area for about 7 or 8 years.

Alex, I would approach your question differently. In a situation where others are scaling back and lack our capabilities, these are the opportunities where we can capitalize on our prime locations and structure them optimally. Consider the minimal amount of land we have invested in currently. Additionally, this environment tends to produce some of our most profitable projects, both in terms of the initial deals we secure and those projects that will launch a couple of years from now, which will face less new competition.

This is Mike on with Haendel at Mizuho. My question is, does the D.C. DOGE job cuts multiplier effect on the D&B region give you less confidence in market rent growth going into 2026? And how does that potentially impact the acquisition property and your portfolio in that region?

Mike, this is Sean. Two things. One is in terms of the ripple effect, as I mentioned earlier, the DOGE impact, if anything, is probably being felt around now given the lag effect between the time people were noticed and when they actually departed. If there was a ripple effect, we probably would be seeing more of that now. I think it's a little uncertain at this point that we've actually seen that. And then on your second question, we have not acquired anything in this region in quite a long time in terms of assets. Is there another question there, Mike?

This is Ami. Are the remaining deliveries and lease-ups from the supply cycle more focused on urban or suburban areas? If interest rates continue to decrease and development activity resumes, do you think the focus will shift more towards urban or suburban locations?

Ami, it's Matt. Looking ahead over the next year, we anticipate more deliveries will come from urban submarkets compared to suburban ones. There are a couple of exceptions, specifically in Northern California and possibly New York. However, in most other regions, we expect a slight increase in supply for both urban and suburban areas, and the difference between them is narrowing. It was wider two years ago and slightly wider this year, but I’m somewhat surprised that urban supply is still forthcoming. As we look further ahead for new developments, the economics favor suburban areas right now. This seems to be the general market trend. However, obtaining entitlements in established suburban regions is challenging, so it often takes time to prepare a deal. Additionally, we are observing that many urban centers are now promoting the conversion of outdated office spaces into multifamily units and offering incentives for this process. If this trend gains traction, we could see those supplies become available relatively quickly, as these are existing buildings that can be converted with a shorter construction timeline.

Just a quick one for me. We saw the spread between renewals and new move-ins widen again this quarter, and part of that's due to the seasonal trend this time of year. But curious if you have any thoughts on that dynamic and what that means for rent growth for both front-end pricing and renewals moving into '26?

Yes, Alex, it's Sean. Yes, fair point. I mean, typically, you would start to see a seasonal shift between the rent change for renewals versus move-ins at this time of the year. And nothing new on that front. Other than on the move-in side, as I indicated in my prepared remarks, it has been weaker on the move-in side in terms of rent change than we would have anticipated, reflecting some of the deceleration that we've talked about in some of the markets that I identified earlier like the Mid-Atlantic and L.A. and Denver. So certainly a little more meaningful than seasonal in those particular markets, but you would expect that to continue likely through year-end. And you don't start to see any kind of shift in that in a material way until you get to the spring leasing season and asking rents really start to move up through that period of time typically. So that's what would normally occur. At this point, we haven't provided a forecast for 2026, but that would follow the historical norm.

Thank you, everyone, for joining us today, and we look forward to connecting with you soon and at NAREIT in early December.