Good day, and welcome to Essex Property Trust Third Quarter 2019 Earnings Conference Call. As a reminder, today's conference is being recorded. Statements made on this conference call regarding expected operating results and other future events are forward-looking statements that involve risks and uncertainties. Forward-looking statements are made based on current expectations, assumptions and beliefs as well as information available to the company at this time. A number of factors could cause actual results to differ materially from those anticipated. Further information about these risks can be found in the company's filings with the SEC. It is now my pleasure to introduce your host, Mr. Mike Schall, President and Chief Executive Officer for Essex Property Trust. Thank you, Mr. Schall. You may begin.

Thank you, operator, and welcome, everyone, to our third quarter earnings conference call. John Burkart and Angela Kleiman will follow me with comments before we open the call to Q&A. I will begin by highlighting that our third quarter results exceeded our expectations. Core FFO per share for the quarter was $0.04 better than the midpoint of our guidance, which represents 6.3% growth compared to the third quarter of 2018. We are pleased to raise our full year core FFO guidance by $0.05 at the midpoint. This is the third time this year we've increased our core FFO per share guidance, which is attributable to a solid summer leasing season, dramatically improved cost of capital compared to the start of the year and continued execution by the Essex team. John and Angela will provide more details about the quarter and the increase to our full year guidance in their remarks. Turning to our expectations for 2020 rent growth in the West Coast markets. We have continued our practice of providing our baseline and key budgeting assumptions on Page S-16 of the supplemental. This is intended to be a scenario based on macro level forecasts, such as U.S. GDP and job growth that we obtained from third-party sources. Key to this scenario is our ground-up research on apartment supply throughout the Essex metros. In recent years, apartment supply has become more concentrated in California's urban areas, resulting in periods of high rental concessions that often substantially impact submarket pricing power. Therefore, apartment supply estimates are more important than ever, and our delay adjusted supply estimates for 2019 have proven invaluable in targeting acquisitions and dispositions. Overall, we expect 2020 to look much like 2019, representing the continuation of rental growth near long-term averages. Overall, we expect market rents in the Essex…

Thank you, Mike. Q3 was a good quarter for Essex. Overall, our markets follow the historical seasonal pattern with the market rents peaking in early August, a little later than normal. As a result, we continue to favor achieving market rents over higher occupancy, taking advantage of the strong market conditions to lock in higher rents into September, allowing our same-store portfolio occupancy to dip to 95.8% or 60 basis points below September 2018. This strategy enabled us to increase year-over-year scheduled rent by 3.5% for the third quarter compared to the prior year's period. For comparison, year-over-year scheduled rent growth in the third quarter of 2018 was 2.5%. This 100 basis point increase sets us up in a better position going forward. Our total same-store market rents were up 3.4% year-over-year in the third quarter over the prior year's period. Revenues for the same period grew 3.1% with financial occupancy of 96%, which is 40 basis points lower than the prior year's period. On a trailing 12-month basis, portfolio turnover rate was 46.3% through September. This is a 200 basis point decrease from the prior year's period. Finally, we continue to be on pace to achieve the 3.3% midpoint of our full year 2019 same-store revenue guidance. On supply, multifamily supply growth as a percent of stock is projected to be 120 basis points higher in downtown urban submarkets during 2020 at 2% versus suburban markets at 80 basis points. Concentrated downtown urban supply is a consistent theme in this cycle in the West Coast markets. Looking forward, we continue to be excited about the various platform initiatives that we are working on, including our new CRM that we're collaboratively designing with a third-party vendor, our mobile maintenance platform 2.0 as well as our smart units and many other…

Thank you, John. I will start with a brief overview of our third quarter results and the increase to our full year guidance and provide an update on our structured finance investments and balance sheet activities. Beginning with our third quarter performance. I'm pleased to report that we exceeded the midpoint of our core FFO per share guidance by $0.04. This was primarily driven by investment activities and low interest expense, including favorable refinancing and higher capitalized interest from development delays. Following these results, we are raising our full year core FFO per share guidance by $0.05 to $13.33 at the midpoint, representing a 6% year-over-year growth. The revised guidance range assumes we sell a large urban asset in Northern California in the fourth quarter, which is owned in our co-investment platform at a 55% pro rata share. The sale of this property will take us to the low end of our full year disposition guidance range in the low $300 million. Turning to our structured finance investment activities. The recent decline in interest rates in a strong transaction market have enabled developers to either refinance or sell their development projects. As a result, there has been an increase in early redemptions of our preferred equity and subordinated loan investments. During the third quarter, we have $31 million of early redemptions, and we currently anticipate an additional $110 million of redemptions in the fourth quarter. As we maintain an active pipeline, we plan to reinvest the proceeds from these early redemptions and achieve comparable yields. But until the reinvested proceeds are fully funded, there will be an impact to core FFO on a temporary basis. In addition, during the fourth quarter, our investment in a collateralized mortgage obligation, which was originated back in 2010 for $70 million, will mature. As a result, we expect to receive a payment of approximately $80 million. This investment has been highly profitable for our shareholders, earning around a 470% return over a whole period. It is another good example of our opportunistic approach to investing. However, replicating this investment in today's low environment, low rate environment will be challenging and will lead to an estimated FFO per share headwind of $0.14 to $0.18 in 2020, subject to reinvestment yield and timing. Lastly, on capital markets activity, we also took advantage of the decline in interest rates to issue $550 million of unsecured bonds in the third and fourth quarter with a coupon of 3%. The majority of this debt is being used to repay secured debt maturing in 2020. We will continue to be opportunistic as we consider our refinancing alternatives to optimize our cost of capital. This concludes my remarks, and I will now turn the call over to the operator for Q&A. Thank you.

[Operator Instructions] Our first question is from Austin Wurschmidt with KeyBanc Capital Markets.

First question is really around Prop. 10, 2.0. And I guess, since the initial Prop. 10 vote in 2018, other than AB 1482 getting passed, I was just curious what else has changed in your mind that could potentially swing the vote in 2020 in your view.

Okay, Austin, it's Mike. Yes, we certainly hope that we would not be following up on Prop. 10 this quickly, but here we are. So to the contrary, actually, we are hoping that the bills passed would be given time to work. And so anyway, what's happening currently is signature gathering is ongoing for the proposition. They're somewhere around 600,000 signatures. They probably need around 900,000 signatures in order to get rid of the duplicates and unregistered voters, et cetera. So they still have a ways to go. As you know, we've kept our Californian for Responsible Housing entity in place that was used as part of the Prop. 10 battle, and we're just starting to focus on the proposal. But I would say, remember that Costa Hawkins Prop. 10 was overwhelmingly defeated in 2018, a lost in all but 1 county in California. It was a decisive victory. And again, I think the politics are a little bit different this time in that the, I think the Governor and legislature would like have some time for the existing bills to work before going to the ballot box. So I think it's still too early to tell exactly what's going to happen, but we are definitely focused on it. And we're organized and ready to fight it as needed.

But would you say anything sentiment-wise from the voters that either changes to the proposed bill or something amongst, I guess, just voters in general that they're taking a different view because we all know that it obviously was voted down pretty significantly in 2018?

Yes. So I think that there have been some changes that make it more palatable to voters, but again, less palatable in another sense. So it exempts individuals that own 2 homes or less, for example, from the provisions. But then again, there is a vacancy control element, which I think will be viewed as being undesirable. So I think that there are offsetting pieces there. And again, I think the politics are very different this time because, as you'll recall, with respect to Prop. 10, the Democrat party endorsed it. I think that's very unlikely this time. So we still think that it doesn't get supported and will be voted down. But again, it's a little bit early to come to that conclusion, I guess.

Okay. Appreciate that. And then I know you said there's not a material impact from AB 1482. Was just wondering if you could put a finer point on that and what the impact to 2019 same-store revenue growth would have been if we were looking at that being implemented Jan 1 of this year.

Yes, this is John. The, again, from Essex's perspective, for years, we have done a couple of things. One, we self-imposed rent caps, which are relatively consistent with 1482. And number two is we typically, we always focus on the market and we try to bring the renewals slightly below or at market. So we're not really ahead of the market with our renewals. So with that said and the fact that rents are growing roughly 3% a year, it probably would not have had a material effect on '19 and nor will it on '20. The types of issues that you can run into is if you're pushing ahead of the market or, so there is a little bit of friction as it relates to the term. Sometimes, shorter-term leases get some price premiums, but we have relatively few of those. So I don't see it having really a material impact on our business, but we'll obviously include whatever that will be in our guidance when we issue that next quarter.

Our next question is from John Kim with BMO Capital Markets.

I was wondering on the CPI with 5% rent cap, what do you think that will have as far as impacts on your renewal rates and your same-store results?

Yes. John, I think John will probably want to add to this. We run our renewals and new leases, new lease rates at pretty similar numbers for the September '19 year-over-year. New was 3.1% and renewals 4% overall for the company. So because of that, again, as John said, we don't see it having a huge impact on our portfolio. John, do you want to add to that?

Yes. No, I mean that answers that. Again, it's the way we operate, it's, again, consistent with how we've operated for years with self-imposed caps, with how we look at the market and how we look at our renewals with respect to the marketplace. So yes, again, I do not see a material impact. There may be minor impacts as it relates to terms and how we adjust, and we're working through all that right now, but it's just really not a material impact.

Okay. And then my second question is on now being that acquirer with the cost of capital getting more attractive, are there any major underwriting assumptions that you're doing differently this time that you, that you're taking now versus before? And can you update us on where you see the best return as far as acquisitions, preferred investments and development?

Sure, John. This is Mike. In terms of underwriting, we look at things much like we have in the past. We will look at every transaction based on the submarket rent growth. And again, our research department ranks our submarkets by 4-year rent growth divided into 30 submarkets. And so we're trying to invest in the top performing submarkets from a rent growth perspective over that period of time. So that's how we do that. And we'll call the portfolio with respect to the bottom part of that hierarchy. And in terms of best overall performance, I think that on acquisitions, we look at it relative to the portfolio. So in other words, what are the returns that are available in the private markets versus what other, what returns are implied in the stock and try to understand how we create value, how we can create core FFO per share and NAV per share growth. Both of those metrics are the key to what we, how we look at an acquisition and then, of course, whether we use on balance sheet sources of money versus our co-investment program. So all these factors go into how we underwrite a transaction and how we approach it and really no different from before.

Our next question is from Trent Trujilio with Scotiabank.

On your 2020, I guess, preliminary outlook there, the job growth forecast for Southern California looks to be basically a continuation of what we've seen year-to-date. And now as you look at your markets and think about you seem to do a deep dive into office space requirements, tech growth, VC funding and other factors, would you say that Southern California job growth has a higher likelihood to surprise the upside or downside than what you put in that forecast?

It's a good question. I wish I have that answer. And it would, if I did have that answer, it might influence what we, what our decision-making is, but we tend to gravitate. At this point in time, if you look at the top part of our portfolio from a growth perspective, it still remains the North and the tech markets. Although definitely, Southern California is betting, benefiting from some of the tech companies moving operations to Southern California. I think Seattle is benefiting more from that trend out of the Bay Area, but Southern California definitely will get some advantage. As in Southern California, it's been in this low 1% range for several years. And so to bet that it's going to move a lot is probably unlikely, I would guess. It's a huge, it's the, the largest part of our portfolio. It's a huge amount of people and jobs. And so it's hard to make that work. It's a more diversified economy, more reflective of the United States. And so practically speaking, I think that the job forecast is not going to vary by a whole lot. If you look at this past year, we said, originally, Southern California a year, going to a year ago, the '19 forecast. Southern California would be at 1.3%, and we had Northern California at 2%. So we've done quite a bit better in Northern California. It's currently at about 2.7%, and Seattle has done a lot better. So it's really been driven by the tech markets. And so I think our view is ride that tech wave. And we think Northern California and Seattle are the best way to do that.

Appreciate the color on that. And then looking at your development schedule, I noticed that Station Park Green Phase III looks, it's nearly complete and yet construction cost went up about $10 million or $58,000 a unit versus last quarter. Can you talk about that a little bit more and maybe provide some context to why that is and how that's perhaps affecting your view of future development starts?

Well, yes. We, I've commented actually in the script and many times before about this difficulty of direct development in that construction costs are growing much faster than rents, and it's compressing development yields. And as you see us go from Phase 3 to Phase 4, you see the cost go up pretty significantly. Again, as it relates to Phase III, specifically, it's caused by shortages of labor and longer periods that the building that's under construction and similar type problems. And so again, this is, I'd like to say that we're different from the comments that we make about direct development, but we are part of that world. And so you're seeing the outcome.

Our next question is from Shirley Wu with Bank of America.

So my first question has to do with the supply that you've seen in '19. So in your '19 forecast, you guys forecasted around 35,700 units coming online in '19. So I was curious about the slippage that you do anticipate going into '20 and what submarkets those were, are probably going to be in.

Yes, Shirley. This is Mike. Yes, there are some nuances here. As to '19, we were actually very close. If we go back to our original expectations for 2019, we're still very close to that number. So I give our research department great credit for doing the, such a good job building the supply pipeline up and being pretty accurate. There was, there have been some delays that were a little bit greater than what is implied on our forecast, but it's really been something that has added a lot of value. What will change between '19 and '20 is the sort of the cadence of the supply delivery. So in 2019, in Q1 and Q2, we had roughly 16,000 units delivered out of the 35,000 total. In 2020, that will go to about 19,000. So we're actually projecting that Q4 2019 is the peak on a quarterly basis to supply. So even though it looks like it's the same number year-over-year, the reality is that it's peaking now and will decline throughout 2020. This will have an impact on pricing in that we will have more supply deliveries in the first and second quarter, which, obviously, are more impactful to our 2020 guidance. And, but again, as we've noted on, in our presentation at the BAML meeting, overall, we hit the peak on building permits in a couple, a year or 2 ago. And so we're now down about 13% from that number. So we think that we get the peak of supply as to permits. We think that we'd hit the peak as to deliveries in Q4 2019, and it looks like we get better as the year goes on in 2020.

Great. So then I guess, just to follow-up on that. Your strategy going into the early '20 would be to keep occupancy and to pull back a little bit on rates then.

Our strategy going forward, you're saying. Yes, as I mentioned in my remarks upfront, our portfolio right now is at a 97.2% occupancy, and that was very purposeful. We held out, benefiting from the stronger peak leasing season and the peak in rents, which was later this year, again, another good thing that we benefited from. But then we rapidly repositioned the portfolio, recognizing the supply that's coming at us. So yes, we will hold out higher supply, but, I'm sorry, higher occupancy, but more than that, we positioned our portfolio, which will enable us to avoid some of the lows. When I say choppy, what I mean is, occasionally, people start pricing units down too aggressively. And if we're in a good position like we are in right now, we'll have the ability to hold back, wait a week and hold back and let things clear up and then lease our units. So I do expect our occupancy to come down some from the 97.2%, but we'll generally favor occupancy for the next couple of quarters, for sure.

Got it. So my next question is on expenses. So year-to-date, your expenses are up 3%., but you maintained expenses at 2% to 2.4%. So I guess, how should we think about the rest of the year with just the different buckets in expenses and admin and R&M was a little bit higher for the quarter as well?

Yes. Shirley, it's Angela here. On the expenses side, it's tough to look at quarterly run rate and then try to assume what that means because expenses are, by nature, very lumpy. And you'll see that last year, our expenses ran higher in the second half, particularly in the fourth quarter, whereas this year, it's a little bit more throughout the year. And when we set our expense planning with the team, what we don't do is we don't focus on when they're going to execute certain expense items. It's really whenever it makes sense for them to do so and if there's a good window to do so. And so at the end of the day, you'll see that lumpiness and it's probably going to continue. Having said that, we are expecting to come in, in our guidance range, and we are on plan as far as the expenses are concerned.

And our next question is from Nick Joseph with Citigroup.

I appreciate the color on the operating strategy. And sorry, if I missed it, but what's the current loss to lease for the entire portfolio versus this time last year?

Sure. Let me start. The loss to lease is obviously a function of the market rent versus scheduled rent. And so as I mentioned, we had a phenomenal year this year, increasing our scheduled rent 3.5%. So we took a lot of the rent that was available in the marketplace, the 0.5%. So we took a lot of the rent that was available in the marketplace. And then we also held out longer for adjusting our portfolio, which meant, that said, our numbers are 90 basis points for September of '19 versus 150 basis points for September of '18. But again, because of these adjustments, a better way to look at it is really recognizing that the, if we average August and September together for both years, they come up actually equal. So we're about in the same spot as we were last year, but we drove more money down to the bottom line because of the market conditions and how we played it. Does that answer your question?

It does. That's very helpful. And then Mike, maybe just on the transaction market. For assets broadly that are impacted by 1482, what do you think the cap rate impact could be going forward?

It's a good question, Nick. For sure. I don't think it's going to have a huge impact because as long as CPI plus 5% is somewhere around 8%., the market is not underwriting 8% rent growth. And so you may have some years that you're coming out of a recession, for example, where you might be constrained by those current parameters. But I think the marketplace is underwriting longer-term rents. I know everyone on the call looks at things much shorter term, the markets take the longer view. And you don't get anywhere near CPI plus 5% in terms of long-term rent growth. So I think it will have relatively small impact on transaction values.

Our next question is from Rich Hill with Morgan Stanley.

I wanted to come back to the supply comments just for a moment and really ask 2 questions. First, and foremost, just so I'm clear, it sounds like supply has just been pushed out a little bit into 2H. 4Q '19 is the peak and that will still have some ripple effects through 1Q and 2Q. But once we get past that, if I'm thinking about this correctly, we should finally be past the supply. Am I thinking about that correctly?

This is Mike, and we don't have our numbers for 2021 yet, so flying blind a little bit, but I definitely agree with what you're suggesting. We think that the peaks have been had with respect to permits, and we're starting to see it on the deliveries. And so I think it will level off at a level that will be quite a bit below the 35,000, but there's still plenty of interest in development. So we're not saying it's going to a nominal amount. We're just suggesting it will trend down. And I think it continues into 2021.

Yes. Helpful. And I want to come back to your comment about plenty of interest in development. And Mike, I just want to play devil's advocate here for a second. You've noted that interest rates have really helped your acquisition volumes, and there's a tremendous amount of negative yielding, fixed income, debt globally, by our count, around $15 trillion. Is there any scenario where development yield, or do development suddenly begins to reaccelerate because development yields actually look really compelling on a relative basis, particularly in markets like San Francisco?

Yes. I mean, definitely, I think that, that could happen. It tends not to happen later on in the economic cycle because you have things that pull against you. Labor shortages cause construction cost increases to be very large. And that is a top-of-the-market phenomena, generally now bottom of the market phenomena. For your assets and rents are at pretty high levels. So compare it to the 2012 to 2016 period where we were growing same-property revenue, an average of 7% per year, that's not likely to happen at the top of the cycle given the constraint of rent to income and affordability. And so we try to triangulate this stuff pretty carefully. And what we've thought is, as I said in the prepared remarks, preferred equity is a better way to approach that because we're coming in late in the process. We're coming in at the point the construction begins. The costs are known. The bank financing is in place, et cetera. And similar to one of the deals that we did a couple of years ago, Century Towers, where we would take both an ownership position in a development deal, as well as a preferred equity piece. So we would consider doing that. But the point is we want to come in late in the cycle and, because that's where you get better growth and cities generally are more agreeable to helping make development work. And just the whole scenario is better.

Our next question is from Rob Stevenson with Janney Montgomery Scott.

Mike, of your 3 acquisitions this quarter, it looks like 2 JVs and 1 wholly owned. How do you guys think about how much more aggressive you can bid for deals in a 50% JV with the various fees and promotes than you'd be comfortable bidding for a wholly owned asset? I mean, does the 50% JV structure allow you to bid 25 basis points lower, 50 basis points? How meaningful is that gap when you're able to bring the co-invest program to bear?

Yes, it's a good question. Actually, the, we have a partner on the other side. And if we're going to pursue something, which will generally be the larger transactions in a joint venture mode. We're working with them on these scenarios. We generally have a sort of an internal benchmark sensitivity analysis as to whether it's more beneficial for us to be in a JV format or on balance sheet. And essentially, over time, we go back and forth in terms of, depending upon the cost of capital and the impacts on core FFO per share and NAV per share. So we're constantly changing the mix. And, but I would say, we're not trying to push the pricing because if we start pushing the pricing, guess what, we start affecting future deals. And so we want to buy properties that we think are well suited, well located at an attractive yield, a yield that works, it adds value both from a core FFO and an NAV per share basis and not influence the market. So that's our basic program. It really hasn't changed a lot. And I cringe a little bit about the thought of, oh gee, we should bid more because we can bid more. We really try to avoid that.

But there has to be some sort of gap there, right? I mean whatever it is, because, otherwise, at this point in the cycle, given your $20-some-billion entity and opportunities are more scarcely capital for you, you do them all on balance sheet, right?

Well, no, not really, actually. So in the sensitivity analysis that I was talking about, we need a certain premium to NAV in order to utilize our balance sheet as opposed to a joint venture transaction. So just working through the math on these things. And so right now, our preference, strong preference is to use joint ventures. You will notice that we did do one deal on balance sheet, but it was a relatively small deal township, and it, that was driven more by the size of the deal rather than its overall impact. So that's how we look at it. Again, it's not in our best interest, I don't think, to try to push the market cap rates down. We want cap rates to remain as high as possible. So, and then it's really the other way around. Our transaction people try to find high-quality deals in markets that are going to grow and then we try to find the best capital to fund that deal. So the transaction comes first. The capital decision comes second.

Okay. And then second question. Year-to-date, average monthly rental growth in the same-store portfolio was 3.4%. How much are you seeing expansion in fee growth in the same-store portfolio, pets, parking, trash, et cetera, and how we sort of reached the point, not only in your portfolio, but it is an industry when you're sort of maxed out on the fees, you can charge the residents that you're already charging $4,500 a month for rent?

Sure. This is John. The fee part of it is running just slightly in front of the base rental revenue at this point in time. But that said, there's a level of change that's going on within the different line items. For example, you have cable and you have the cable cutting, so that's going down. At the same time, parking is going up. And then there's few other line items that are moving, where pets is fairly flat. So I think longer term, there'll be some opportunities. I had mentioned a while ago, we were working with, renting some of our amenity space to nonresidents. And at the same time, I commented that immediately, 2 of our peers picked up, that off the call and with our same vendors. So we're pretty quiet about the details that we have going on but, going forward at least. But I do think there will be some level of opportunity with that. The big thing though always is rents. And Essex has always been focused on our assets, on our locations, on our research, and that will always be the number 1 most important aspect of Essex. But we do try to enhance our returns wherever possible.

And how meaningful are the overall fees or whatever you're calling fees out of the revenue standpoint these days?

You're talking a few percent.

Our next question is from Alexander Goldfarb with Sandler O'Neill.

Mike, on the rent regulation, on AB 1482, and then you got Prop. 10 2.0, I thought that part of 1482 superseded whatever local towns wanted to do. So what would overturning, and maybe that's not correct, but what would overturning Costa-Hawkins do given the implementation of 1482?

Yes, Alex. Actually, 1482 does not affect what current rent control regulations that are in place. So it sits separately from that. So those regulations remain in place.

Right. I didn't mean overrule the current regulations. I thought it governs stuff going forward. Is that not the case?

Well, you have Costa-Hawkins out there, that's a statewide law. You have 1482, that's statewide law. And then you have the local regulations. And maybe I'm not understanding your question. But each of them sit as a separate entity, and they, one does not supersede or overrule the next. They, you have to comply with each of them. Does that make sense?

Yes. No, Mike, that absolutely does. That absolutely does. And then the second thing is, Angela, you mentioned the $0.14 to $0.18 of headwind. And I think, if I heard correctly, that was from the early repayments that you got in third Q and that you're expecting in 4Q. So maybe just a question of how quickly do you think you can put that to work. And then, Mike, you spoke about the competition or what's going on in your preference for the developer equity. But clearly, others, your peers are getting into that business. Maybe you can talk a little bit about how you guys view a deal for income versus underwrite a deal that you may actually want to own the asset at the end.

Alex, on the structural finance and, comments, let me just make sure I clarify. So on the early redemption of, that's coming toward us in Q4 of $110 million, that does not relate to the $0.14 to $0.18 headwind for next year because we do expect to backfill those prefer equity investments. So the question really is there will be some headwind because as these deals get backfilled, they don't fund immediately. They tend to take up to 6 months to fund. And so that's one item, and I don't have a specific number for that at this point in time. But I highlighted the 14 to 16, I mean, $0.14 to $0.18 because that's more meaningful from a single investment, and that relates to the CMO that we originated back in 2010. So it's separate from the preferred equity. And that's a onetime item for that instrument.

And then, Alex, I'll try to hit the second part of your question. So when we realized that there definitely is more competition for preferred equity, that's definitely a true statement. We feel pretty good about the relationships that we have formed, and a number of our relationships have come back for a second transaction. And so we feel pretty good, and we feel like we have an advantage there relative to others. We're a known quantity. We've been in the business a long time. We have great construction lending relationships that we can marry with our preferred equity, et cetera. And we've always tried, and I've commented on this on the call, too, to leverage those preferred equity investments into other types of relationships, either by trying to get an option to buy at the end. And we generally don't get an option to buy primarily because the developer thinks it's worth a lot more than we do at the end of the construction period. So that's been a headwind. So you end up, you would take a much lower yield and then have an average transaction at the end. That's not attractive to us. We do have a seat at the table. There have been several of these transactions that we've either purchased at the end or purchased shortly after construction completion and/or converted to a longer-term, lower preferred interest that keeps us in the capital structure. Again, as I said in my comments, one of the, there are a couple of opportunities to do some direct development, and this would be one of them, which is marrying our preferred equity with a direct ownership interest in the deal, where we're coming in, again, later on in the process. So we continue to look for ways to leverage the preferred equity into other types of transactions. And for whatever reason, this whole series of refinances, I think we looked at a couple of those deals, refinances and sales Angela referred to in the comments, looked at some of those deals. But a number of them, we did not, we just went on and sold the property and/or we didn't want them. So again, I think that preferred equity keeps us in the hunt from a variety of perspectives, and it's worthwhile and it's also great financially.

Our next question is from Steve Sakwa with Evercore ISI.

Really just one question, Mike. As you sort of think about AB 1482 and some of the other potential legislative changes, has it sort of altered your investment strategy as it relates to submarkets, age of assets, types of buildings that you're willing to buy? And conversely, is it impacting how you think about the portfolio and what you might want to sell going forward?

Yes. Steve, good to hear your voice. It does, to some extent, I guess, beginning at the, a couple of years ago when this risk was increasing, looking at our portfolio from the perspective of the cities that have really horrible rent control, rent control that basically shuts down new development. It was essentially fixed by Costa-Hawkins in that Costa-Hawkins prohibited pernicious forms of rent control to older property in effect. So I'd say we have been more cautious with respect to those cities. They'd be basically San Francisco, Berkeley, Santa Monica, et cetera. But beyond that, I think that the discussion in California has become a pretty balanced discussion in that it's not just about protecting the renter, which is obviously important, and it's why 1482 was a bill that the California Apartment Association did not oppose, but at the same time, don't change the ability or the attractiveness of developing more housing because there's an enormous shortage in California, estimated by one study at 3.5 million homes and probably getting much worse. That was as of 2015. It's probably gotten worse since then. And maybe another way of looking at it is, I think, in this cycle, we produced about 5x more jobs than homes created in the Essex markets. So 270,000 jobs to 55,000 homes. So the housing shortage is getting worse, not better. And the last thing we need is to shut down the development pipeline. So I guess we view 1482 as a balanced proposal. We view the amendment of Costa-Hawkins as essentially an anti-growth, very unbalanced proposal. And so that's how we look at it. So I guess it would change it in a couple of very specific areas in terms of where we want to own property, but it hasn't had a huge impact on the overall company.

Our next question is from Hardik Goel with Zelman & Associates.

I've got 2 for you today. The first one, on supply, just to understand your thinking here. I acknowledge permits are down 13% over the last 12 months in our markets. But if I look at the pipeline of units that have been started, that pipeline, whenever it's slated to be delivered, if I just add up all those that have started, that's 5x what's been completed over the last 12 months. So even if I assume like a really high abandonment rate or I assume a lot of it is delayed, it seems like it's a multiyear supply pressure kind of thing that's maybe either in late 2021. Is that consistent with what you're seeing? How would you think about that?

Yes. No, it's not at all consistent with what we're seeing. I mean here's the problem. I mean the data vendors have a very difficult time trying to weed out which projects are shown 3 different times: by name, by location, by, and you really need to sort through that data and really understand where it is in the process. I know there are a lot of metros in the U.S. that you can take a permit and/or a start and you can go out 24 months and you have an apartment building. It just doesn't work out here. And so you really have to be involved in a more fundamental level, driving the assets, understanding what they look like, where they are in the development process in order to get a good estimate what's going to be produced. So, because for 2020, if it's not under construction right now, it's not going to deliver in 2020. So I think we know, again, 50 units and more is what we do, but we send our people out and they drive the buildings because we view supply as being a critical thing. Supply used to be more scattered around, throughout our marketplaces. It's become much more concentrated. And as a result, we have to understand it much better than we have in the past. And so that's why we spend a lot of time on it, getting that right. So we feel very comfortable in our 2020 numbers. And yes, sorry about the basic data that's out there. It's tough to follow.

No, that's really helpful. And I completely agree to 2020 kind of outlook. It's just hard to see through to 2021. And just to follow up on what you just said, what does that pipeline number look like in terms of, I mean you said it takes so much longer to build in California. That must mean that if you look at what's under construction right now, you can see through the visibility is higher for 2021 as well, right?

Yes, you're right. Yes, you are right. We just haven't focused on 2021 yet. So we'll get there. We'll be talking about 2021 next year. I don't have that information right now. I'm sure it exists within Essex. I just don't have it right now. And I suspect that, again, as you go through 2020, you have a pretty significant drop-off. You start with 10,000 units in Q1 and you have somewhere around 7,000, 8,000 units in Q4. So that drop-off is definitely there. But I just don't have the next steps in front of me, and I don't want to make it up.

Sure. And just lastly, if you could share the new and renewal for the quarter, just those spreads.

Sure. This is John. So for the quarter, our new leases were, on average, the, here we go, sorry, 3.4% and the renewals going out in the fourth quarter, 4.5% on average. The actual achieved for the third quarter was 4.2%.

Our next question is from John Pawlowski with Green Street Advisors.

Mike, the acquisitions this year have been focused in Northern California. Can we expect the Northern California bias to continue next year if your cost of capital stays advantageous?

It's a good question, John. The pipeline right now is pretty heavily weighted in Northern California and for, maybe to add a little bit of color to that, we see this rent-to-income ratio at about 103%. So incomes have done really well in Northern California, and so that becomes less of a constraint. And Southern California has not really kept up to that number. But we hope that we can also buy in Seattle. So those 2 markets would be our focus. But we might buy an asset or 2 in Southern California as well.

Okay. John, could you help us quantify a bit the choppiness you're referring to in Northern California that you expect over the next few quarters and just be a bit more specific? If, so for instance, if Northern California today, on average, is printing 3.5% revenue growth, if choppiness hits, what does that trajectory of revenue growth look like this time next year?

Sure. So the choppiness really relates to what's going on with market rents at any point in time, not necessarily revenue. There's a difference, obviously. That, in part, was why strategically we filled the portfolio up to 97.2% and are in a good position going forward. So what I'm expecting the, as I mentioned, in Northern California specifically, the supply is twice as much in the second half of this year, '19, as it was in the first half. And unfortunately, that's coming at a time when demand is seasonally slower or lower. So that will impact market rents, and it will largely emanate around where the supply is being delivered. So you can, you can imagine, in places like downtown Oakland, where there's a little bit more supply, there will be an impact. But in the sense of revenues, we're obviously not giving out next year's guidance, but we're reaffirming this year's guidance. And it really, I don't see it having a huge impact on us because we positioned ourselves well to avoid that. I think there'll be some headlines of rents being down. But again, I use the word choppy because it's not that they're going to be down long term. You got to step back for a second and realize the Bay Area is a very strong market, has very strong job growth. There's supply that's coming into it. Big picture is under 1%. So we're in a good spot. It's just going to create some headlines of lower rents for the quarter and probably into the first quarter of next year, and that's what I'm really referring to.

Okay. Understood. Just a follow-up there. One submarket that stood out within that broader healthy backdrop this quarter was San Mateo and just in terms of the sequential trends. Anything specific there on the demand side that you're seeing outside of lumpy supply hitting?

Yes. No, that one is really a market that the results there for the quarter for San Mateo are really a result of the market being strong. And what I mean by that, actually, it's a very strong market. And I think our own ops team got over there. Skis on that a little bit. And so we had a vacancy decline of over 200 basis points in that market, which is what drove that number. But that's obviously resolved. So it was really the function of the market being so strong. They got, I think, a little bit overconfident with that particular market.

Our next question is from Rich Anderson with SMBC.

15 minutes before the call starts, this is what I get. An hour and 10 minutes in. So thanks for sticking around. But I do want to ask a couple of, so when I think of the Essex perhaps of old, you're always kind of at least a couple of hundred basis points better than the national average from a same-store growth perspective. And you're kind of more, by your standard, more pedestrian type of internal growth today. Is it all entirely this concentration of supply issue? Or are there other dynamics, perhaps on the demand side, that have changed such that kind of a return of Essex as being this premium growth provider are perhaps gone for a while?

It's a good question, Rich. Appreciate it. I'd say every economy, every recovery period is a little bit different. And so there is no one size fits all. And there will be times when we, I think, pretty significantly outperform. And this business is one, I think reasonable people would agree, that rents and incomes have to grow pretty close to one another over long periods of time. And so, but this cycle, I think, is somewhat different. And I said earlier, from 2012 to 2016, rents grew. Our same-property revenue grew 7% a year. And that was much faster than incomes. And therefore, we created an affordability issue, which we are now mending. And it is getting better. So we see, I don't think anything has changed. I think we are in the markets that generate the highest CAGRs of rent growth over long periods of time. And certainly, this cycle seems to be pretty strong. We got a lot more, a lot better rent growth for a long period of time, which I think everyone has forgotten. But it's set up this affordability issue, and we're working through that. And you know what, we've made progress, a lot of progress. So from, as I mentioned before, from 2010 to 2015, rents outgrew incomes by about 220%. If you look at that number from 2010 to 2019, rents outgrew income by about 30%. So it's getting tighter. And we are solving that affordability problem, and there will be a point down the road where rents will do better.

Okay. That's great color. And then very early on in the call, you mentioned how supply is really concentrated in the urban areas, which you generally are not. And I'm curious if you've been able to quantify how well Essex has done relative to your urban counterparts from an internal growth perspective. Is there a number that you track to see how you're doing purely on that, using that supply observation?

Yes. This is Mike again. I think that there are just fundamental differences between us and the competitive set. And I think one of the key differences is that we, our focus is core FFO per share growth and NAV per share growth. It really is not tied to the same-property metrics that are quarter-to-quarter, et cetera. And why does that matter? Why would it disconnect? Well, it would disconnect to the extent that you make investments in property or you overinvest in property, for example, which use capital dollars. And yes, it will push up your same-property revenue growth, but it doesn't actually add value on a bottom line basis. So when you look at large capital investments, capital investments that have relatively short duration income, we're less excited about that. We are interested when we do renovation projects, for example. We're looking for a long-term return consistent with real estate. And there's lots of things that can be done, investing in personal property, shorter-term, revenue-generating type proposals that can make those numbers look better. That's not our focus. Our focus is on growing income and core FFO per share, cash flow over long periods of time. And I think, go look back at the chart, Rich. There, I think the proof is in the detail. There's actually one chart on our, in our presentation that, I think, sums it up, which is we produced a long-term, 25-year CAGR. We've been a public company for 25 years. The CAGR of shareholder return is 16%. I don't think there's a lot of companies that have done that over 25 years.

Our next question is from Neil Malkin with Capital One Securities.

I just have 2 questions and 5 follow-ups for each of those questions. The first one is I saw that your notes receivable increased by about $145 million this quarter. Is that a change in classification of that remix that I think you were talking about selling? Or is that something else? Some color on that would be helpful because that's a pretty large amount.

The notes receivable increase on the income line is really because we have 2 bridge loans associated with the acquisitions that we provided to our joint venture. And so those are, once again, temporary in nature, but they do generate some income for a couple of quarters of their outstanding until we secure permanent financing.

Okay. I know that a lot of questions have been asked about supply, but some, I'm trying to dig into the data. It looks like the East Bay, San Jose, 2 of your largest markets, supply is expected to, as a percentage of existing stock, should go up by more than double. And even if you assume this double counting phenomenon you're talking about, that would imply that like almost all the products are double counted. So I guess, is it just a gross error in the vendor, like Axiometrics, for example, that is massively overstating the impact? Or any, again, any help there, trying to understand because to me, it seems like you're setting up for a situation like we were a couple of years ago when the Bay Area had rents that fell precipitously due to clusters of supply relative to job growth that was not of the early cycle?

Yes. I might preface the history with I think you're referring to probably 2015, '16, when the Bay Area slowed pretty dramatically. But again, still very positive. And back to my prior comment that if rents get too far out ahead of incomes, it creates an affordability issue, which is like a constraint. It isn't a supply-demand issue. It's a constraint issue. But people make different decisions like doubling up, tripling up, whatever. And so that becomes a factor. Again, back to this rent to income in Northern California, it's currently at 103% of the long-term historical average, which makes it more affordable than Southern Cal at 109%. And so we think that Northern Cal is essentially fixing that affordability issue faster than everyone else. Our numbers, and again, I can't comment on that, as to the data vendor. And we realize it's challenging to get accurate data with respect to supply. Having said that, as I said previously, we were really accurate in 2019. I give our people a lot of credit for doing the hard work of putting this together. But if I just look big picture at 2020, we expect in Northern California about 72,000 jobs and we are producing about 18,000 total housing units. So 11,000 apartments and 6,400 single-family homes. That relationship is 2:1. So candidly, we need more housing in Northern California, and it doesn't appear that there's a supply-demand issue with respect to that. I mean you would, so I think it looks good. 72,000 jobs should translate to about, demand for about 36,000 homes, total homes, not apartments, so total. And we produce about 18,000. So that still seems to work as far as we're concerned.

I hear you on that. I just, like I said, I mean, some of the data providers, they have addresses, names, a lot of details. So I mean it just seems like it's odd that they'd be off by such a large amount. So maybe that's something I can look into more.

Well, maybe just go back, go back a year and look at what the projections were and then look at what they are now. I mean I think that's probably the best way to do it.

All right. And then last one. Short-term rentals are becoming a little bit more talked about. I just wondered, just given the, especially your types of markets, maybe a higher transient or migration-related demographic. Are you, how much, if any, are you looking to put that into your portfolio? Is that something that you're seeing more demand for, more interest in? And if so, what kind of role do you see that playing in your portfolio?

Well, this is John. We are, of course, always looking to figure out how we can maximize the value of the assets. And so we do explore different opportunities and evaluate the impact. What we see oftentimes on, if you're meeting like the overnight type rentals, we see that, that has a negative impact on the quality of life for our customers or other customers and ultimately hurts our overall market, hurts our market rents. At the end of the day, it creates issues there. Having, you can imagine having people coming in with all their overnight stuff through the leasing offices and into the amenity space. And they have a different frame of mind, a very short-term frame of mind as opposed to the tenants that live there, and it's their home. So we continue to look for opportunities. And there are some situations where that's working, where people are taking a floor or a building and segregating it that way to basically define different spaces. But there are still issues then with the amenities crossing over. So we'll constantly look at that, but I think it's not a huge opportunity at this point in time unless something changes.

And this does conclude our question-and-answer session. I would like to turn the call back over to management for closing remarks.

Thank you, operator. And thank you, everyone, for participating on the call today. We hope to see many of you at NAREIT in a few weeks. Have a great day. Thanks.

Thank you. This does conclude today's conference. You may disconnect your lines at this time, and thank you for your participation.