Good morning, ladies and gentlemen. Welcome to the AvalonBay Communities' fourth quarter 2019 earnings conference call. At this time, all participants are in a listen-only mode. Following remarks by the company, we will conduct a question and answer session. You may enter the question and answer queue at any time during this call by pressing star one. If your question has been answered or you wish to remove yourself from the queue, press star two. If you are using a speakerphone, please lift the handset before asking your question, and we ask that you refrain from typing and have your cell phones turned off during the question and answer session. Today's conference is being recorded. Your host for today's conference call is Mr. Jason Reilly, Vice President of Investor Relations. Mr. Reilly, you may begin your conference.
Thank you, Nadia, and welcome to AvalonBay Communities' fourth quarter 2019 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's 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, this press release does include an attachment with definitions of reconciliations to non-GAAP financial measures and other terms which may be used in today's discussion. The attachment is also available on our website at www.avalonbay.com/earnings, and we encourage you to refer to this information during your review of our operating results and financial performance. With that, I'll turn the call over to Tim Naughton, Chairman and Chief Executive Officer of AvalonBay Communities, for his remarks.
Well, thank you, Jason, and welcome to our fourth quarter call. With me today are Kevin O'Shea, Sean Breslin, and Matt Birenbaum. Sean, Kevin, and I will provide some comments on the slides that we posted last night, and then all of us will be available for Q&A afterwards. Our comments will focus on providing a summary of Q4 and the full-year results, and a discussion of our outlook for 2020. Starting on slide four, highlights for the quarter of the year include Core FFO growth of 5.2% in Q4 and 3.8% for the full year. Same-store revenue growth came in at 2.5% for the quarter and 2.8% when you include redevelopment. Full-year same-store revenue growth came in at 2.9% or 3.1% including redevelopment.
We completed $335 million of new development in Q4 and $665 million for the year at a 6.5% yield and started just under $800 million last year. Lastly, we raised $1.3 billion in external capital for the year in a mix of asset sales, new debt, net of redemptions, and common equity at an average initial cost of 4.3%. The next few slides will provide a little more detail on 2019 performance and provide context for our 2020 outlook. Turning first to slide five. During the year, we saw the East Coast surpass the West Coast in rent growth for the first time in eight years. In Q4, the East outpaced the West by over 100 basis points, led by Boston and the D.C. metro area, while we experienced some softening conditions in California. Turning now to slide six.
Development continued to be a big contributor to NAV growth in 2019. Development completions of $665 million created approximately $300 million in incremental value, and at 30% margins remained very healthy 10 years into the expansion. Moving now to slides seven and eight. We made good progress in Southeast Florida and Denver over the last year, our expansion markets. To date, we've committed roughly $600 million to each market through a combination of acquisitions, development, and partnering with local developers. These portfolios are comprised of new assets, and as you can see from the maps on slides seven and eight, are spread across a broad geography within the markets. Turning now to slide nine. We believe to be a great company, we need to take a multi-stakeholder approach to our business.
We can only deliver strong financial results over a sustained period of time by having engaged associates, satisfied customers, and the support of local communities by taking an active leadership role in addressing important environmental and social challenges. While we always strive to be better, we've been recognized for doing a good job in many of these areas over the past year. Starting with our associates, as engagement scores are in the top decile, and we were named to Glassdoor's list of top 100 best places to work for the second consecutive year in 2019. With our customers, as we ranked number one nationally among apartment REITs for online reputation for the fourth consecutive year.
Lastly, with our communities, where our efforts on the ESG front have been recognized by several organizations, including GRESB, who recognized AVB as the U.S. and global leader in the residential sector, by the Carbon Disclosure Project, or CDP, an international organization that grades companies on their carbon emission disclosure practices across all industries, and who recently awarded AVB a grade of A-, one of four REITs and the only apartment company to receive such a grade. By the Science-Based Targets initiative, who reviewed and approved AVB's submittal for targeted reduction in carbon emissions by 2030. We are one of 11 real estate companies globally and one of five REITs that have completed this process. Lastly, by CR Magazine, who ranked AVB in the top 100 corporate citizens globally for the second consecutive year out of a universe of over 1,000 companies. Turning now to slide 10.
Now that the calendar has turned to 2020, we thought it made sense to look back on our performance over the last decade. In short, it's been a great decade and cycle for the apartment sector and for AvalonBay. Buoyed by strong demand fundamentals and attractive capital market conditions, we've seen healthy growth in earnings and NAV since 2010. For AVB, Core FFO growth has averaged 10% on a compounded annual basis since 2010, or about 200 basis points above the sector. The absolute level of growth is fueled by both strong internal growth, as same-store revenue growth has averaged over 4%. As can be seen on slide 11, by strong external growth, mainly from contributions to new development, as we consistently delivered new development at initial yields well above the marginal cost of capital and roughly 200 basis points above prevailing cap rates.
As you can see on the right-hand side of the slide, development deliveries since 2010 of just over $8 billion have generated more than $3 billion in initial value creation or roughly $26 per share. Turning now to our outlook for 2020 on slide 12. We're projecting Core FFO growth of just over 5% for this year, mainly from a combination of same-store NOI growth of 3% and from stabilized new development, which is expected to generate more than $60 million in NOI in 2020. Turning to slide 13 and double-clicking on this a bit. The stabilized portfolio is expected to contribute roughly 3.5% of total Core FFO growth in 2020. External growth or NOI from new investment net of capital costs is expected to contribute 2.7%.
Overhead growth and loss of JV and fund income will actually provide a drag of about 110 basis points to growth, with about 1/3 of that 110 basis points coming from a loss of JV and fund fee income, about a third from normal overhead growth in our base business, and the remaining third coming from expense overhead that represents investment in innovation or other overhead that is projected to generate future ROI, some of which we will benefit from in 2020. At 5.1%, the level of Core FFO growth is expected to be 130 basis points higher than 2019. The biggest driver for this is stronger external growth of roughly 200 basis points on a net basis, driven by higher levels of deliveries in late 2019 and 2020, as well as a lower cost of capital raised last year.
This is offset in part by the headwind from the loss of JV and fund income as we wind these vehicles down. Turning to slide 14. This slide provides a summary of the key economic factors driving much of our outlook. I won't go through everything here. Simply say, while the GDP and job growth is expected to moderate from 2019, the strength of the consumer should provide a tailwind with tight labor markets, rising wages, healthy balance sheets bolstering confidence, healthy consumption, and household formation. The business sector is expected to provide more crosswinds in 2020 versus 2019, as sluggish business investment should be offset in part by less concern surrounding a potential trade war. Lastly, the government sector should bolster the economy in 2020, as the lack of fiscal constraint and an accommodative Fed should provide added support to economic growth.
Overall, it's shaping up as a good macro backdrop to support the economy and the housing sector for the year ahead. I'll now turn it over to Sean, who will provide more color on our outlook for the portfolio.
Okay. Thanks, Tim. Just turning to slide 15. This chart represents the actual 2019 and projected 2020 components of total personal income growth, both for the U.S. and for our markets. As Tim noted, the consensus forecast is for decelerating job growth throughout 2020, in part due to continued structural factors constraining growth in the labor force. Consumers, however, are feeling reasonably confident given the low 3% wage growth they experienced in 2019 and the outlook for even stronger wage growth in 2020, which supports healthy demand for our business. Turning to slide 16. New supply for our market footprint is projected to be consistent with what we experienced during 2019, with increases in urban Boston, Northern New Jersey, Los Angeles, and across the Northern California markets being offset by reductions in the Mid-Atlantic, Pacific Northwest, and in New York City.
Turning to slide 17, we expect same-store revenue growth of 2.7% for 2020. On a quarterly basis, revenue growth is projected to be strongest in the first quarter of the year due to a number of unique factors, but should be relatively stable throughout the balance of the year. On the East Coast, we expect a material improvement in performance in the Boston and Mid-Atlantic regions, both supported by a decline in the pace of new deliveries in our sub-markets. Results in Metro New York and New Jersey, however, will continue to be weighed down by the impact of the New York rent regulations adopted in mid-2019. Shifting to the West Coast, we expect continued healthy performance in the Pacific Northwest. Job growth exceeded 3% in 2019. Consensus expectations reflect a job growth rate that's roughly 1.5x the national average in 2020.
Relatively strong demand, combined with a modest reduction in the pace of new deliveries, should continue to produce greater than 3% revenue growth in the region. Turning to California, we expect decelerating performance in both regions. In Northern California, job growth is projected to decelerate from roughly 2.6% in 2019 to 1% in 2020. In addition, new supply is projected to increase by roughly 800 units in San Francisco and between 1,000 and 1,200 units in each of San Jose and the East Bay, weighing on performance throughout the year. In Southern California, job growth is projected to fall from 1.6% in 2019 to about 80 basis points in 2020. On the supply side, while new deliveries are expected to be relatively flat year-over-year in Orange County and San Diego, deliveries in L.A. are projected to increase by roughly 1,500 units as compared to 2019.
From a broader perspective, we continue to expect a headwind from New York rent regulations that were adopted during mid-2019 and the implementation of Assembly Bill 1482 in California. The impact of these two rent regulations on our full year 2020 rent yield revenue growth rate is estimated to be between 15 and 20 basis points. Turning to slide 18, our outlook reflects sub 2% same-store operating expense growth in 2020, with increases in various controllable and non-controllable expense categories partially offset by a reduction in payroll costs. The reduction in payroll expense is driven by our innovation efforts, particularly those related to reimagining the leasing experience for our customers and associates. As I mentioned last quarter, elements of the redesigned experience include the use of an AI-powered automated agent for lead management purposes and a more dynamic demand-driven staffing model.
Additional elements that are underway and projected to yield a benefit in 2020 include more self-guided tours and an express move-in process. Through year-end 2019, we have reduced our on-site sales and management staff by approximately 7.5% relative to our baseline staffing in 2018 and expect to stabilize at a roughly 10%-12% reduction later this year. Of our total increase in same-store operating expenses, roughly 60% is related to growth in property taxes and insurance. Given the timing of tax appeals and supplemental assessments in various markets, along with the timing of certain other items in various expense categories, we expect operating expense growth to be front weighted in 2020, as noted in the text box on slide 18. With that summary of our markets and outlook for the portfolio, I'll turn it over to Kevin to talk about our development activity and the balance sheet.
Thanks, Sean. Turning to slide 19, since the middle of the last decade, when new development starts averaged $1.4 billion per year, we've reduced starts to about $800 million per year. This shift was driven by both a reduced opportunity set for attractive development and by a desire to limit development activity to an amount we could fund on a leverage-neutral basis without needing to access the equity market. For 2020, we expect to start about $900 million in new development on projects that we expect will not only improve the quality of our portfolio, but also generate compelling value creation in line with recent completions. Turning to slide 20, as Sean has just outlined, we continue to innovate across our operating platform, investing in initiatives that will allow us to better serve our customer while generating higher returns for our shareholders.
For 2020, our investment in these initiatives is expected to account for roughly half the increase in expense overhead costs, adjusted for non-core items. Turning to slide 21, we show our $4.2 billion pipeline of future development opportunities, which are controlled at a very modest cost and offer a lot of flexibility as it relates to the timing of the start of construction. About half of our development rights are conventional conditional agreements or options to purchase land with private third-party land sellers. The remaining half is split between asset densification opportunities, where we are pursuing additional density of existing stabilized assets in public-private partnerships, which are generally long-term development efforts that span a number of years. These types of projects allow more flexibility to align the start of construction with favorable market conditions.
In addition, it's worth noting that in creating this pipeline, we've been careful to limit our financial exposure so that we enjoy an attractive set of development opportunities at a modest upfront cost. At the end of the fourth quarter, land held for development was zero, an all-time low, and pursuit costs represented an additional $70 million. This allows us to control over $4 billion of future development across our markets for an upfront investment of less than 2% of projected total capital costs. Turning to slide 22, as we've discussed before, another way in which we mitigate risk from development is by substantially match funding development underway with long-term capital. This allows us to lock in development profit and reduce development exposure to future changes in capital costs.
As you can see in the slide, we are approximately 85% match funded against the development underway at the end of the fourth quarter of 2019. On slide 23, we show several of our key credit metrics and compare these to the multifamily REIT sector average. As you can see, our credit metrics remain strong in both absolute and relative terms, reflecting our superior financial flexibility. Specifically at year-end, net debt to core EBITDA was low at 4.6x , unencumbered NOI was high at 93%, and the weighted average years to maturity of our total debt outstanding remained high at about nine years. With that, I'll turn it back to Tim.
All right, great. Thanks, Kevin. In summary, just on slide 24, 2019 played out more or less as expected. A good year for the apartment sector. We delivered 3%+ NOI growth. We completed $665 million of accreted development, and we made good progress in growing our presence in our expansion markets. In 2020, we expect the economy and apartment markets to remain healthy. For AVB, we're expecting same-store NOI growth of 3%. External growth will be stronger than what we experienced in 2019 due to higher levels of deliveries and lower capital costs. Lastly, we plan to continue to operate the business, as Kevin mentioned, in a risk-measured way with respect to capital formation and deployment, recognizing that we are 10 years into the current expansion. With that, Nadia, we'll open up the call for questions.
Thank you. If you'd like to ask a question, once again, please signal by pressing star one on your telephone keypad. If you are using a speakerphone, please make sure your mute function is turned off to allow your signal to reach our equipment. We'll first go with Nick Joseph from Citi. Please go ahead.
Thanks. Sean, you talked about the innovation efforts. What do you estimate the total opportunity to be in terms of either NOI or margin? How much will you realize in 2020 versus future years?
Yeah, Nick, good question. What I'd probably point you back to, and happy to talk about it further, is the comments we provided last quarter. We went through several slides and talked about roughly a 50-basis-point improvement in margins for each of two primary activities. One, on the, what we call the leasing journeys, leasing and move-in process combined. On the maintenance side of the house, about another 50 basis points related to efficiency opportunities in that bucket as well. In terms of what we are going to see in 2020, I would describe it this way. We started this back in late 2018. As we move through 2020, our run rate by the end of 2020, which is associated with the leasing side of this, which is really what's paying off in 2020, will be somewhere in the range of $7 million-$8 million.
We'll be bleeding in the maintenance activities later this year. You really won't see as much of that benefit until subsequent years. There are additional components to the leasing side that will also bleed in in subsequent years. You'll see more as it moves into 2021, 2022, but as it relates to 2020, the aggregate impact that's flowing through represents the two years of activity and roughly a $7 million-$8 million benefit.
Thanks. That's helpful. Just on the same-store revenue growth. In 2019, you had 20 basis points benefit from redevelopment. I know a lot of your peers include that in their same-store guidance. For 2020, what do you expect that benefit to be from redevelopment activity?
Yeah. On rental revenue, we expect it to be about 15 basis points, and on NOI, about 20.
Great. Thank you.
Yep.
We'll next go with Rich Hightower from Evercore. Please go ahead.
Good afternoon, guys. I'm just looking at the market-by-market chart for same-store revenue. Just as you look at New York, what would it take to get to the high end of the range this year, that 3% number, given what you're baking in with respect to the rent regs and maybe as an add-on to that, you can maybe talk about what the new broker fee rules mean, if anything, for your portfolio or just the landscape in New York? Thank you.
Yeah, Rich, it's Sean. The fundamental issue is just better overall rent change as we move through the year. If you look at what we're anticipating in the metro New York, New Jersey region in 2020 versus 2019, is like-term lease rent change coming down around 40, 50 basis points. Depending on what happens with various market conditions and particularly the suburban markets for us, is what's going to move the needle. Depending on how things manifest itself with job growth, if things are stronger, we might see a benefit there, but we're trying to reflect what the macro assumptions are for the economic environment there, the constraints imposed by the loss of fees that will continue to bleed through. The fees are known.
Really, it's on the demand side and pushing more rental rates through the suburban portfolio for the most part that would push those numbers up. Yeah. On the broker fees, that's really not material for us. Our portfolio there, we really haven't paid brokers in quite some time. We have a handful of the JV assets that have paid a few, but we have paid those fees as opposed to pushing them back onto the residents. From what you see from a total reporting standpoint, it's immaterial. It's basically no different for us.
Okay. Yeah, that's what I thought. Just wanted to confirm that. Maybe with respect to capital sources guidance for the year, maybe a question for Kevin.
I guess outside of condo sales, which you do disclose, can you just break down that bucket, the $1.4 between asset sales and sort of other capital markets, whether other equity or debt? Just help us understand where the most favorable cost of capital is from that perspective right now, or how you guys think about it.
Sure, Rich. This is Kevin. Just a few things. As you saw from the release, we anticipate sourcing external capital of about $1.4 billion in 2020. I guess the first thing to point out is we expect to do so on a roughly leverage-neutral basis relative to net debt to EBITDA, where we're at 4.6x at the end of 2019. Roughly flat to that in terms of how we bring in the mix of capital. As you know, there's three markets we tend to look at, unsecured debt markets, the asset sale market, and the equity markets. Here, we do have condo sales happening that are in the mix.
Our capital plan for $1.4 billion, we don't typically break it down in detail, but when you model it out, just bear in mind, we tend to be leverage neutral, and we do not have equity in the plan. It's a combination of debt, mostly unsecured debt, overall dispositions, and condo sales, which you can see from our earnings release. We expect about $240 million in condo sales in this year.
Got it. Thank you.
Yeah.
Thank you. We'll next go with Richard Hill from Morgan Stanley. Please go ahead.
Hi, guys. This is Lauren Weston on for Richard Hill. Appreciate your comments on expense guidance. You could just help us understand, this is the lowest guidance we've seen in recent history. I just want to get a sense if this is a new run rate or really particular to 2020. Obviously, the focus is on payrolls coming down sequentially. Can you just give us more color on sort of what we should expect going forward beyond 2020?
Hello, this is Sean. Providing expense guidance well beyond 2020 is speculative at best. What I would say is that the big categories to really think about, without giving sort of precise answers in terms of what to think about for 2021 and beyond is, property taxes, more than a third of our expense structure. If you look at that, I think values are growing at a pace that's a little bit above inflation, but you don't have the same kind of acceleration based on the significant run-up in values that we've seen this cycle. You should start to see that level off for us. In addition, we don't have really a large portfolio of 421-a assets or other PILOT assets that are bleeding in at a very high rate. That will help contain the overall expense growth in property taxes.
The other big piece is payroll. You kind of know what's happening with wage growth and payroll, which is sort of in the low- to mid-3% range. Our job is to try to innovate as best we can to contain that. -1.5% is not a sustainable run rate as you push things through, but we would expect to contain it at something below inflationary levels for wage growth based on the innovation that we expect over the next few years. The other ones really kind of the other, like R&M and some of those things really sort of grow at inflationary levels for the most part. Those are kind of the big pieces and the way I think about each one.
Okay. That's helpful. Then a follow-up, if I can. Just in the Pacific Northwest, a little bit of a spike there. I saw a footnote in the supplement, but any other color you can add on the appeal process there and the success there?
That change in expenses in Pacific Northwest was driven by successful appeals actually in the fourth quarter in the prior period. As you may know, early last year, I think all of us benefited from a reduction in rates throughout the various counties within which we operate across the Greater Seattle, Pacific Northwest region. We also had the benefit of appeals that were realized in the fourth quarter of 2018, which put upward pressure on the growth rate in Q4 2019.
Thanks.
Thank you. Once again, if you'd like to ask a question, signal by pressing star one . If you'd like to be removed from the queue, please signal by pressing star two. We'll next go with Alua Askarbek from Bank of America. Please go ahead.
Hi, guys. Thank you for taking the call. Just a little bit about Seattle further. It appears that Seattle's continued to weaken. Can we get any extra color there in terms of the supply impact and when you expect the bulk of it to pass with supply coming down next year?
Yeah. As it relates to Seattle, some commentary just on our performance there in the fourth quarter. A bit of sort of unusual one-time things coming through that depressed our growth rate in the quarter down to 2.2%. Our actual base rental revenue growth rate, we kind of just think about the residential component, was 3.1%. As a result of a number of sort of one-time things in the other rental revenue category, we lost about 90 basis points. That relates to payments we received for revenue share in the prior year, as well as a one-time thing in terms of cancellation fees that didn't come through in December and came through in January. I would say that the overall performance in the Pacific Northwest is quite healthy. It's probably the strongest in the portfolio right now in terms of rent change.
January rent change across Seattle was almost 5%, and revenue growth in January is popping back up and should be in about the mid-3, mid-3.5% range or so. I wouldn't extrapolate from the fourth quarter performance into 2020. We still expect it to be a very healthy market this year.
Got it. Thank you for that. Just a little bit on the supply one. What are you thinking about it for the rest of the year? When do you think most of the supply will peak for that?
In terms of Seattle specifically?
Yes.
Seattle specifically, we're expecting to come down in 2020. The pace of deliveries on a quarter-by-quarter basis is pretty evenly spread. It's 1,800-2,000 units a quarter or so. There's a little bit more Q1, but for the most part, it's not materially different in terms of the pace of deliveries throughout the year.
Okay, great. Thank you.
Thank you. We'll next go with John Kim from BMO Capital Markets.
Can you comment on what you're seeing as far as development yields in Southeast Florida and Denver versus your legacy markets?
Sure, John, this is Matt. I can comment on that a little bit. The first thing I'd say is it's a little bit of an apple and an orange because we have not obviously been in those markets for the same length of time with the same kind of depth of experience. We do expect going in that, for us anyway, the yields will be a little bit lower at first. That hopefully over time, as we establish our position there, we'll be able to post even stronger returns. In Florida, we're really partnering with other developers there. The yield to us is thinner because we are not taking the construction risk or the entitlement risk. We're really just taking the lease-up risk on those deals.
We're looking at kind of low-five yields for us on those types of deals, where there's a sponsor who's taking a significant amount of the value spread, if you will. In Denver, we just started the deal last quarter in RiNo. That is our first ground-up development in Denver, that is not with a partner. That was a buy right deal in the city of Denver. Pretty predictable process in terms of getting our permits, that's kind of a high-five yield. I think that's probably what we're seeing right now there, just given that it is a market that's seen a lot of hard cost pressure. I would imagine over time that both of those markets would be in the high fives. Maybe even low sixes in certain cases, probably more likely kind of into high fives.
When do you think you get into enough critical mass to bring management in-house in those markets?
Denver, we actually have brought management in-house. We brought management in-house late last year. We do think that over time, that will be an advantage for us. Florida, we're kind of looking at that now. I think that's in the plans for later this year.
Okay. Just a follow-up on the New York announcement banning broker fees. I realize maybe not a direct impact to you, but I'm wondering what you think about how this impacts the market and if you subscribe to the view that this will lead to higher rents in New York.
Yeah. This is Sean, John. Hard to say. The cost has been there. I think the question is how does it affect sort of landlord behavior? Whether they continue to engage brokers in the same way or not. There are a lot of people that do, and there are a lot of people that don't. I think it's kind of behavioral at this point in terms of how it's going to impact things. I mean, the demand is the fundamental demand to find a unit to occupy. The question is, who's paying the transaction cost or if there is one. It's too hard to tell in terms of where that's going to come out.
What about market share? Is there a chance that you're in a better position than some of your peers?
I don't know about that. I think certainly us and our other REIT peers, in terms of the sophisticated systems we have around digital marketing and penetration efforts there and other sources of demand, are probably more well-positioned in the short run. In the long run, the market will correct to reflect the dynamics of the regulatory environment. In the long run, it may be a part, but I'd say those with more scale that can push through more digital marketing efforts on average, the margin probably should benefit some, yes.
Very helpful. Thank you.
We'll next go with John Guinee from Stifel. Please go ahead.
Great. Very nice guidance, guys, gentlemen. Thank you. Hey, I was looking at the recent CBRE stats, I think those guys are pretty good. Over the last four years, you've seen completions average about 67,000 a quarter. Over the last two years, you've seen starts about 100,000 or more a quarter. It just seems with those kind of basic numbers, it's only a matter of time before deliveries bounce up over 100 a quarter. Does that make sense to you guys? Have you guys factored that into your medium-term thinking?
Hey, John, it's Tim here. In terms of our markets, starts have actually been pretty flat. Nationally, I mean, our numbers I think have been running more in the mid threes, and our starting finish up to 400. If not, just a little bit more than that. Your numbers don't sync up quite the same as at least the numbers we use to track national starts. In our markets, I think it's been running in the 80,000 or something like that. It's been pretty flat over the last few years, last three years.
All right. Great. Thank you.
Yep.
We'll next go with Nick Yulico from Scotiabank. Please go ahead.
Oh, thanks. In terms of the development starts, I'm just hoping to get a feeling for the yield you expect on the new starts this year, and also maybe a preview on the type of product and markets that you're focused on.
Sure, Nick. It's Matt. Our starts basket for this year obviously can change a little bit, based on how the year shapes out. It's probably a low to mid-60s yield, somewhere in that 63- 64, 65 range. That's really driven a lot by the mix of business. There's one or two deals in Long Island that we will hope to start this year, one or two deals in northern New Jersey, both of those are markets that tend to be higher yield, higher development yield markets. We do have a big deal in L.A. in downtown, in the Arts District, we're looking to start. Then probably one deal each in Denver and Florida. It's a nice mix. On average, more in some of those higher-yield markets.
Okay. That's helpful. Thanks. I just had a question about the condo sales, the gross proceeds this year of $230 million-$250 million. Can you give us a feel for what percentage of the project that is? Is it 40%, 50%? I'm just trying to get a feel for how much proceeds are still left to come back as capital after this year.
Sure. This is Matt. Again, I can speak to that. Just where we stand as of right now today, we have three units that have actually closed. They closed in January. They represented about $10 million in gross proceeds. We have 51 additional executed contracts with deposits up that represent about another $150 million. Of those 51, actually, we expect 43 to close before the end of Q1 in February and March. Most of those proceeds should come in in the next couple of months. We actually have seven additional offers which we've accepted, but they're not yet in contract. A good amount of that is spoken for. In terms of what's actually in contract or closed, it would be those 54 at $160 million. Now, that is not a pro rata share of the building.
It's based on which units happen to be selling and settling first. You can't just take that as a percentage of the unit count. What's left to sell is probably a little more than that pro rata because there's some high-dollar units at the top of the building that are not in that number. That gives you a good sense of where we are.
Okay. Yeah, I guess I was just wondering, I think in the past you guys have said somewhere around $3 million a unit. That would get to right around, say, $500 million of total gross proceeds from the condos, which means you get half of that this year and half of that in 2021. Is that the right way to think about it?
Yeah. We'll see where the market goes from here. Sales pace could be better or faster or slower. Like I said, we know that we have a fair amount of that already kind of spoken for.
Okay, thanks.
Thank you. We'll next go with Drew Babin from Baird. Please go ahead.
Good afternoon. This is Alex on for Drew. We were hoping if you guys could provide color on January's lease in terms of Northern California, and maybe if you guys could just speak to kind of the dynamics you're seeing. I know you referenced kind of a pretty significant deceleration in employment growth, and it looked like there was another significant deceleration in like-term effective rent during the second half of 2019. If there's any color on that market and kind of what you saw in January would be helpful.
Yeah, Alex, this is Sean. A couple things I'd say. One is, Northern California overall, we expect more supply to come in this year. As I noted in my prepared remarks, overall, it's about a 25% increase across the three markets in terms of deliveries. There's definitely some of that coming into play in the first quarter. One of the things that we did, just given what we were seeing in San Francisco and how we think about a couple other markets, is we did try to build occupancy a little bit more than we might otherwise have done in the fourth quarter, particularly late in the fourth quarter. You see that reflected in the rent change that we reported on last night. Northern California was weak. That carried over into January, with blended rent changes about 60 basis points or so.
It is the lowest of the regions in terms of rent change for January. When you look at it in terms of kind of where rents are today, asking rents relative to last year, they're up about 1.5%, which is one of the lowest of the regions at this point in time. It's a little choppy right now with some new deliveries coming in and our expectations for continued deceleration in employment growth. We'll see how it plays out throughout the balance of the year in terms of overall performance. We are expecting about an 80-basis-point reduction in like-term rent change across the region relative to what we actually produced in 2019.
Thanks for the color there. That's really helpful. One more question from me. Can you speak to the Alderwood Mall development? Is there a retail or kind of mixed-use component? Is your partner related to the mall at all? Just kind of curious what went into the underwriting there and if there's more opportunity kind of paired with a retail component or kind of mixed-use.
Sure, this is Matt. Absolutely. Well, that deal is a joint venture with Brookfield Property Partners, the former GGP, who own the mall. It's actually the site of a former Sears box. It is part of the densification play at that mall. There will be retail on the ground floor of the building that we will not own, that the Brookfield will own. We don't necessarily have any exposure to the retail. We do think that this is a growing line of business for us. We're pretty excited about it, and a model that we hope to replicate in a number of other locations. This would be the first that's planned to be a long-term joint venture, where we will own the asset together going forward, and we will earn some fees associated with that.
We actually did also start, one of our other starts this quarter was Woburn in suburban Boston. That's one of these lifestyle center, not a regional mall deals, which is a partnership with Edens. It's actually the second deal we've done with Edens. The first was Mosaic in Northern Virginia, which is eight or nine years old now, seven years old. We actually have a third deal hopefully with Edens coming in the next couple of years as well in Central New Jersey. We do see this as a trend that's growing. It's frankly been a little more difficult to get at it than maybe we had originally thought, particularly on the mall sites, where all of these sites are encumbered with cross-easements, and there's a lot of parties at the table, including sometimes multiple anchors at the mall that have some say in this.
We are making good progress. We have other deals in the pipeline that hopefully will be similar. We do think it's something that plays to our strength as a competitive advantage for us because these are typically not broadly marketed land sites. It's important to the seller, who's our partner, that the quality of the execution and the certainty of the execution, the balance sheet, all the other things we bring to the table matter just as much as just the raw land value.
That's really helpful. Thanks for taking the questions.
Thank you. We'll next go with Richard Skidmore from Goldman Sachs. Please go ahead.
Good afternoon. Just a question on future development, slide 19. As you look at that slide over the next two to three years, do you stay in that kind of range? What might change your view to either go up or down?
Tim here. The $4 billion, we think that could support about $1 billion a year, roughly, in starts, ± $200 million. As Kevin mentioned in his remarks, we have been more in the $800 million range recently, expect to start around $900 million this year. The reality is you can't bring that up too quickly just because of the gestation periods on these deals. It's often two or three years. It would take, for us to really materially increase starts three years from now, we'd have to be starting today, in effect. The reality is just the, other than maybe the mixed-use opportunity that Matt mentioned, the opportunity set isn't as compelling 10 years into the cycle as it was five or six years ago.
Got it. Thank you. Just to follow up, in terms of the future development rates, looks like about 50% of the future development rates are in the metro New York-New Jersey area. Maybe talk a little bit about how you're thinking about New York-New Jersey, given the regulations and other things, and is that how we should think about the focus as you go forward, that you'll be focused a bit more on New York-New Jersey?
No, not necessarily, Richard. This is Matt. If you look at where our starts have actually been in the last couple of years, I think that is probably more representative of what's likely to be the start mix over the next couple of years, where metro New York has been. I don't think we started anything in metro New York last year. It was 20% of our starts in 2018. It might be as much as 40% of our starts this year. Average that over three years, and it's maybe 20%-25% of our start volumes. It is a disproportionate share of our development rights pipeline.
Some of that is because some of those deals are long-gestating entitlement plays, where not all of them may make it through, and we don't necessarily have to invest a whole lot to see if they're going to make it through in some markets, in New Jersey or Long Island, for example. Then there's a couple of very large deals there that may or may not come to fruition in this particular cycle, that we've talked about on some prior calls. There's a couple of very large, high-rise deals in there that really do kind of move that needle a fair amount, that $2 billion, if you will.
Great. Thank you.
Thank you. We'll next go with Derek Johnston from Deutsche Bank. Please go ahead.
To begin, we see the Boston metro area has experienced some solid rental and revenue growth this year, and supply is expected to actually taper in 2020, yet some metrics are pointing towards decelerating rent growth as well. I was wondering to see how your team views AVB's sub-market exposure there.
Yeah, Derek, we didn't hear the first part of the question. Would you mind restating the question clearly, please?
You're coming through very faint.
Oh, sorry. I'm looking at the Boston metro area. We see there is some solid rental and revenue growth, and supply is supposed to taper in 2020, and some of the metrics are actually pointing towards decelerating rent growth. I was kind of seeing what your team views AVB's sub-market exposure there?
No, got it. As I mentioned in my prepared remarks, we are expecting deliveries to increase in urban Boston. If you look at Back Bay, downtown, and sort of adjoining sub-markets, I think it's an additional 1,200, 1,300 units coming into that sub-market. If you look at the suburban sub-markets overall, supply is relatively stable.
Our portfolio, which is primarily suburban, we think will continue to benefit, and that's why we see some acceleration in rental revenue growth in 2020 as compared to what we produced in 2019.
Awesome. Thank you. Looking over at expense forecast for the non-controllable items, the full-year setup really looks quite similar to 2019. Are there any notable upcoming tax items that we should look for this year? Thanks.
Notable? I don't think there's anything notable per se. In terms of the non-controllable components, that represents about 60% of our projected OpEx increase in 2020. In taxes, we're expecting sub 3% growth. There may be some benefits there because there were substantial increases in assessments and rates in New England in the fourth quarter that were not anticipated, that obviously we had to book. We expect some appeals to work their way through the system and benefit us potentially in 2020 or 2021. Taxes, as I mentioned a little bit earlier in response to a question, we think should be relatively stable in terms of the growth rate and the noise you're going to see from quarter to quarter, year in, year out. It's going to be more around the timing of supplemental assessments or appeals or things of that sort.
It's just too early to tell where that's going to come out this year.
Got it. Thank you.
Thank you. We'll next go with John Pawlowski from Green Street Advisors. Please go ahead.
Thanks. Sean, I was hoping you could elaborate more on your comments around the choppiness in San Francisco as January unfolded. I know there's a lot of supply. Are you seeing anything on the ground that suggests demand is easing? Just any comments on concession trends as fourth quarter passed and we're into 2020 now, how that's faring one way or the other?
Yeah, John, I'd say it's probably a little too early to tell. I think ourselves and maybe a couple others sort of anticipated what was coming online and took a slightly more defensive posture late November, December in particular, and that bled into some move-ins in January. It impacted January rent change. There are concessions in the market. Whether you want to call it just a reduction in rent or a concession, there's some softness there. It's hard to parse out if there's a big shift in demand. We're not necessarily seeing that. When we pull the marketing levers, we get more demand. The question is just how much we pull them. I think it's a little too early to tell if it's purely just supply versus demand, but my sense is there's more supply. We'll probably have a better sense by the time we get to the next quarter.
Okay. Understood. I want to, Matt or Tim, want to head back to the suburban New York and New Jersey comments, and the market fundamentals right now are decelerating from already pretty restrained levels. As the development pipeline just naturally increases your exposure to Jersey and suburban New York, are you seeing anything on the ground today that gets you a bit more enthused about the demand backdrop for the next three to five years in these markets? Are you going to let that exposure drift higher as the developments earn in, or will there be some disposition activity to pare that development exposure?
Hey, John, it's Matt. I don't know that our view on the kind of fundamentals about those markets has changed in particular. What we have done and what we will continue to do is try and make sure that our portfolio is balanced, that does sometimes lead to disposition activity. There's certainly been, if you look back over the last five, six years, a lot more of our disposition activity has been concentrated in the New York metro area. Until we did the JV in late 2018, all of that was in the suburbs. We hadn't really sold anything in the city. A lot of that, we just sold an asset in Connecticut that closed in January. We have an asset that we will probably sell here in Central Jersey that may close in the second quarter or late in the first quarter.
That's the way we're thinking about it. Those investments that we've been making have been among the best investments in our entire portfolio. When you look at the total IRR on those investments, they tend to start out at a very high development yield. Usually they actually grow pretty well for the first couple years as they kind of find their footing in the sub-market. It's not the most dynamic market among our footprint, so obviously it doesn't necessarily have the same long-term growth profile. That's how we try and balance that out, is through dispositions, particularly some of the older assets there.
John, just maybe to add to that. I agree with everything Matt said. With respect to the $2.2 billion of development rights in New York, New Jersey, I think Matt mentioned earlier, actually a little over half of it is just two deals. One of which is a densification play on our site, and the other is a public-private deal that is not likely to happen anytime soon, just given it's sort of caught up in politics at the moment. It's really about, it's closer to $1 billion in the metro New York, New Jersey when you kind of correct those two, which is about a third of kind of what remains, which isn't too far off our total allocation to the metro area.
As Matt mentioned, as you've seen, the disposition focus has really been in the Northeast as we recycle that back into development as well as acquisitions in other markets, particularly expansion markets.
Okay. Thanks for the time.
Sure.
Thank you. We'll next go with Wes Golladay from the Royal Bank of Canada. Please go ahead.
Hi, guys. This is Steve for Wes. Just a quick question about cost of capital and insights of development and for the next couple of years. If we stay in a low-yield environment, you're looking at pretty low cost of capital going forward. Are you going to look at your hurdle rates for your next years and the years following the development pipeline and start taking, I guess, what would be lower compared to your historical hurdle rate?
This is Matt. We actually do have a target return matrix that does vary with our cost of capital. There is kind of an automatic, systematic adjustment in our system. As our cost of capital goes up, our target returns go up and vice versa. Actually, if you look at it today, our target returns on new investment are probably as low as they've been in certainly this entire cycle, and that is a reflection of what our cost of capital is today. Now, having said that, it's not a simple, gosh, now we'll take 50 basis points lower on a development than an acquisition yield. We're also looking at total basis. We're also looking at kind of what the risks associated with that are. It's not a simple formula, but it does definitely bear on our thinking a bit.
We've seen that some when you look at some of the more recent starts.
Oh, perfect. That's all I got.
Thank you. We'll next go with Rich Anderson from SMBC. Please go ahead.
Thanks. Good morning. Good afternoon. Excuse me. When I was looking back at this time last year and your guidance, one item stuck out at me. That was your supply or your deliveries estimated for 2019 were 2.3% of stock. They ended the year at 1.8%, and that's your estimate for this coming year. It doesn't appear like the decline from 2.3% at this time last year to the actual that you're seeing today at 1.8% was a function of slippage. Can you explain what the decline came from if there is anything meaningful there?
Yeah. Rich, this is Sean. What I'd say is we've basically adjusted 2020 numbers to reflect our history in terms of the margin of error on delivery schedules. If you looked at it sort of on a growth basis based on what people would tell you, this is what we plan to deliver and when we plan to deliver it, we basically haircut it to reflect sort of a probability-weighted estimate given our history. That's really what's factored into what you're seeing there in terms of the 1.8%. It would've been probably closer to 2%.
Yeah. Right. Yeah. I'm saying, what happened to the 2.3%- 1.8% over the course of the year? Do those projects just turn off, or what happened?
No, they bleed in, so the subsequent period. They'd be bleeding into 2020, but that would've been reflected in the total anticipated deliveries in 2020, that there are other deliveries behind those deals that have now been pushed into a subsequent period.
Hey, Rich, this is Tim. I think this cycle has been a more of a shortage of skilled labor than we've seen in past cycles, and we just haven't experienced this in the past, these kind of delays. I'm not talking about Avalon, I'm talking about the market in general, and it's very common to hear projects being delayed once they start construction by 6-12 months by the time they finish it. It really is a labor issue. I think it's an issue for the country, but it's certainly an issue for our industry.
Okay. I would think the 1.8 would be a higher number that might trend down over the course of the year because of that slippage. I understand.
Yeah. Maybe the shortages are a little less acute today than they were a year ago as well.
Okay. Fair enough. Still on development for you guys, over $60 million or so assumed development NOI in 2020. That's a fairly equivalent number to what you produced in 2017 and 2018. You had a little bit of a hiccup downward in 2019. Is this range, $60 million-ish, sort of like what we should expect on a go-forward basis for you with the 2019 maybe just being an aberration? The reason why I ask is because what you guys have historically done as a company is produced a certain level of same-store growth, but also a number at the FFO line that's much larger than that. Part of that is a function of capital allocation, but also amount of development that comes online.
I'm just trying to get a sense of the cadence going forward because I hate it when I see same-store growth being at or greater than FFO growth.
Yeah. Rich, Tim here. Kevin mentioned in his remarks we've been talking about $800 million-$900 million a year recently at an average. We've been stabilizing in the low 6% range. I think it's just math. That probably gets you closer to $50 million, low 50s going forward rather than $60 million, $65 million, which is where we were kind of towards mid-cycle. We were starting north of a billion a year.
Okay. Last question for me. This time again last year, you predicted your same-store number, and you nailed it pretty much on the nose in terms of the actual result. FFO, the prediction, the guidance actually was lighter than what you actually produced by about 80 or 100 basis points. When you think about the potential to do better over the course of this year, do you think it comes from development, or can it still come from internal sources?
Rich, this is Kevin. In terms of our guidance this time last year for Core FFO, we were at $9.30, and so we ended up at $9.34, so we beat by $0.04. We didn't come in 80 basis points below. Unless there's something else you want, I mean, a clarification, but those are the facts from the Core FFO.
Okay. Maybe I missed all that, but I'll let it go. Thanks very much.
Thank you. We'll next go with Austin Wurschmidt from KeyBanc Capital Markets. Please go ahead.
Hi, good afternoon. Tim, in your presentation, you highlighted the above-average compound annual growth in FFO that you've generated versus peers this cycle. As you shrink the size of the development pipeline, and to the extent that maybe same store, depending on definition, lagging peers at face today, how do you generate above-average FFO growth versus the sector in the coming years?
Yeah, a good question. First of all, the level of alpha was higher for the first five years of that decade than the last five. That, I think, speaks directly to the point where we've had a little bit less external growth in the last couple of years, last two or three years than we did in the prior years. On the issue of same store, I just would really caution you on that one. CapEx is really distorting numbers for the industry right now. You really have to look closely, particularly at that revenue-enhancing or NOI-enhancing CapEx that everyone is spending in a particular year. I think when you sort of cut through it, you're going to find that there's not a big outlier there. Particularly, we're not a big outlier, I can tell you that.
When you look at sort of the market overlap of the seven large incumbents here, there's a lot of overlap. We all kind of have similar exposure. There's not a lot of alpha from what I can tell, from what we can tell. The alphas generally come from external investment activities. Virtually, that full 200 basis points has come from raising and deploying that capital in what we believe is the most accretive way, which has largely been through development, to a lesser extent, redevelopment.
I appreciate that. I know you're below your target leverage today, but would you be willing to operate at higher leverage given the reduced funding risks associated with development and a shrinking pipeline?
Yeah. This is Kevin here. Our general range for leverage is kind of five to six turns on a net debt to EBITDA basis, where we mentioned before, as you're alluding to, 4.6x at the end of the fourth quarter. I think our view is probably 10 years into a cycle, being toward the low end of the five to six range is probably about where we should be. Then targeting if we were to move it up. I think given where we are at 4.6x , our capital plan contemplates kind of being more or less leverage neutral throughout the year. For the right opportunities, I think we have the financial flexibility to obviously increase leverage a little bit, I think probably the upper bound of that would be in the low five turns range.
Thanks. Then just the last one for me, and kind of touching a little bit on what you just mentioned, Tim. What drove your decision to change the definition of stabilized operations for 2020, and could you quantify the impact that it's having on the 2020 same-store revenue guidance? Are you contemplating any additional changes related to redevelopment or revenue-enhancing CapEx in the future?
Yeah. This is Kevin. I'll start here, and Sean may want to chime in. Yeah. What we really did there was just adjusted the same-store occupancy threshold down to 90% from 95%. In doing so, our occupancy threshold is now consistent with the threshold used by five of our six peers. Part of the rationale was just to conform our practice to the prevailing practice within the apartment REIT sector. The other benefit really is that by virtually doing that, it brings more assets from other stabilized and redevelopment into our same-store pool and reduces the size of those other buckets where we don't give guidance to. As a result, more of our overall stabilized and assets are going to be brought within kind of the ambit of our same-store guidance. This should make us a lot easier to model going forward.
Yeah, Austin. In terms of the impact on 2020 guidance, I think I mentioned this previously, but collapsing the buckets together represents about a 15 basis point lift in our revenue growth for 2020.
Anything on any additional changes you're planning on revenue enhancing or redevelopment?
I think you're going to see our revenue-enhancing CapEx drift up a little bit, but still be sort of well below what you see across the peers. To give you some perspective, last three-year average for us, we've invested about $350 a home. If you look at 2017 through 2019, kind of on average, as compared to our peers, I think that number is closer to $5,000. More than 2.5x us. If you put a 10%-12% return on that, kind of back into it, you can see what kind of lift it gives you. We're trying to make sure that, as Tim pointed out, we're making good capital allocation decisions first to development, then to redevelopment, where we think it's accretive, and not just accretive to same store, really value accretive. That's how we think about it.
We're looking for those opportunities in the market to do that in existing assets where we think it makes sense.
Appreciate the thoughts, guys. Thank you.
We'll next go with Hardik Goel from Zelman & Associates. Please go ahead.
Hey, guys. Thanks for taking my question. I know a lot of the questions have been about guidance and the quarter, but I just wanted to take a step back and get your thoughts on something. As I look at the data we've collected from you guys over the last decade or so. We have a lot of data now, and there's a down cycle included in that. D.C. seems to underperform all the other markets quite significantly. I know, Tim, you've talked in the past about how in the long term, D.C. and Northern California might have similar growth rates, and it's a function of volatility in the downturn. It seems like over the last 10 years seems to be a good long-term period.
I just wonder if the apartment REITs in general are missing a trick here with the overallocation of D.C. because of the lack of supply constraints in the market.
Yeah. We see this in other markets too, by the way, that it's had sort of the bad decade, if you will. Certainly, it's true of Boston in the 2000s. Boston was great in the '90s. It was good in the last decade, but not so good in the 2000s. One of the reasons why we try to be in multiple markets, you can't always predict what's going to have the best decade. Seattle certainly outperformed what we had expected it to do this past decade. We didn't expect it to be able to absorb a particular level of supply. It is probably less supply-constrained than D.C., and yet had a great run. We don't think there's really been a secular change with respect to D.C. We think in some ways, the Amazon HQ search was kind of informative.
When you think of ultimately where they landed, it is a great knowledge center that happens to be on the East Coast. You're seeing more and more tech companies sort of diversify their employment base to include D.C. and New York. We want to be overallocated and overindexed to the knowledge economy, and those are largely the markets we're in and some of the expansion markets. A couple of expansion markets we've gone into and perhaps a couple other markets down the road that we've kind of talked about. We're really sort of focused post on demand side, recognizing supply more or less, it catches up a little more quickly in places like Seattle and D.C. than it does in New York and California. It tends to catch up ultimately a bit with demand.
It's really kind of that income and job growth that ultimately is sort of a big driver of ultimate performance. We're good with D.C. If anything, we're probably more bullish on it today than we were, say, three or four years ago in terms of investing the marginal dollar.
Got it. That's really helpful. That's all for me.
Thank you.
Thank you. Once again, if you'd like to ask a question, please signal by pressing star one on your telephone keypad. We'll next go with Daniel Santos from Piper Sandler. Please go ahead.
Oh, hey, good afternoon. It's actually Alexander Goldfarb on for Dan. Two questions. First, Tim, can you just talk about same-store and how the developments contribute to that? Meaning, you guys are 3% this year, which is sort of industry standard, and would think that new development, once it's part of the same-store pool, would have faster growth or maybe it doesn't as it ages until it develops a price point below the stuff that's delivered after, so it can once again have pricing power. Maybe you can just talk about how the development interplays with same-store NOI growth.
Yeah, Alex, this is Sean. If I could take a stab at that one, then Tim and Matt may want to jump in as well. One thing to keep in mind here is, given the size and scale of the portfolio today, development can move the needle, but it's not moving it materially, I would say, in the average year based on the number of deliveries that actually occur in a community. A development stabilizes, it sits in our other stabilized bucket for a year, then it comes in the same store. We typically see a modest benefit from those communities when they come in the same store.
They do actually tend to perform well the first two or three years after they stabilize because the pricing, whether it's rents or net effective rents after concessions, this bucket is trying to lease up the entire community typically within 12 months as opposed to the turnover, which is closer to half the community. Those communities tend to do well, and they typically give us a very modest lift. Given the size of the portfolio today, as it moves in the same store, it's not hugely material unless we happen to have one year where there's a very, very large slug of deliveries. Therefore, when it comes in the same store, it's abnormally large.
Okay. The second question is, you guys, you've been talking for some time about reducing development. You have some severance that's in your numbers this year. I think you had severance last year. Can you just give a sense of two things, where you think your development program will ultimately shake out? The second is, how much cost have you cut? Meaning in aggregate, how much have you reduced? As you reduce the development, does that mean you have increased costs on the P&L or all the reductions are purely costs that were formerly capitalized and that will not end up becoming expensed on the P&L?
Yeah. Alex, Tim, I can start. We have cut back in terms of the capitalized overhead for sure as we've reduced our development volume. We're running, I think, just a little under 3% in terms of development overhead relative to annualized development overhead relative to starts. That includes people's long-term compensation. You sort of compare that to the private world. The private world, they tend to get 3% development fees, if you will, and that doesn't include sort of their promotes or their long-term comp. It's still a pretty efficient model. We are very focused. Honestly, the business units are very focused on this because the way they're incentivized is based upon production and profitability relative to basically FTEs or overhead expenditure. It is something that tends to sort of self-regulate, and it's not exactly linear, but directionally it kind of works.
In terms of whether any of it actually is visible on the P&L, I think it's just in sort of limited-
Severances.
Severances. In limited certain situations where you may be starting up in a region and you don't yet have new deals to capitalize somebody against. There may be some expenses for some period of time, but generally, on a stable region, you're not going to have development expenses sort of being crushed through the P&L.
Tim, do you still see yourself having a few more years of reducing development? In total, how much cost in total have you reduced over the past few years as you've done the severance and shrunk the capitalized overhead?
I might have to get to you offline on the second piece. I think as we talked about in the last couple of years, we really moved from about $1.3 billion, $1.4 billion to about $800 million-$900 million. We wanted to get to a level, I think Kevin mentioned this in his prepared remarks. We wanted to get to a level at which we felt we could fund it without having to be dependent on the equity markets between asset sales, free cash flow, and additional debt. I would say until there's some kind of economic contraction, you should expect we're going to be kind of at this rate. The development pipeline that we have in place should allow us to continue to start about $800 million-$900 million a year.
Thank you.
On the second piece, we'll just get back to you offline on that.
I appreciate that.
Sure.
We'll next go with Haendel St. Juste from Mizuho. Please go ahead.
My question's been answered. Thank you.
This concludes today's question and answer session. At this time, I'd like to turn the conference back to Mr. Tim Naughton. Please go ahead.
Oh, well, thank you, Nadia. Thanks, everybody, for being on today, and we look forward to seeing you in the coming months. Take care.
This concludes today's call. Thank you all for your participation. You may now go ahead and disconnect.