COLD - NYSE

Greetings and welcome to the Americold Realty Trust Third Quarter 2023 Earnings Call. At this time, all participants are in a listen-only mode. A brief question-and-answer session will follow the formal presentation. [Operator Instructions]. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Scott Henderson, Chief Investment Officer of Americold. Thank you. You may begin.

Good afternoon. Thank you for joining us today for Americold Realty Trust's third quarter 2023 earnings conference call. In addition to the press release distributed this afternoon, we have filed a supplemental package with additional detail on our results, which is available in the Investor Relations section on our website at www.americold.com. This afternoon's conference call is hosted by Americold's Chief Executive Officer, George Chappelle; Chief Commercial Officer, Rob Chambers; and Chief Financial Officer, Marc Smernoff. Management will make some prepared comments, after which we will open up the call to your questions. On today's call, management's prepared remarks may contain forward-looking statements. Forward-looking statements address matters that are subject to risks and uncertainties that may cause actual results to differ from those discussed today. A number of factors could cause actual results to differ materially from those anticipated. Forward-looking statements are based on current expectations, assumptions and beliefs, as well as information available to us at this time and speak only as of the date they are made, and management undertakes no obligation to update publicly any of them in light of new information or future events. During this call, we will discuss certain non-GAAP financial measures, including core EBITDA and AFFO. Full definitions of these non-GAAP financial measures and reconciliations to the comparable GAAP financial measures are contained in the supplemental information package available on the company's website. Now I will turn the call over to George.

Thank you, Scott. And thank you all for joining our third quarter 2023 earnings conference call. This afternoon, I will discuss some key operational metrics and financial results for the third quarter, and then comment on our outlook for the remainder of the year. Rob will provide an update on our recent customer initiatives and an update on our growth activity. Marc will also provide some additional commentary on our recent capital markets activity, our third quarter results and a detailed walk through of our guidance for the remainder of the year. Turning to our core business priorities. First, customer service continues to support strong occupancy in our portfolio. For the third quarter, our same-store economic occupancy increased to 84%, which is a 345 basis point increase over last year and an 80 basis point decline sequentially from the second quarter. As we discussed, we expected the sequential decline to be 100 basis points to 200 basis points, but we're able to partially overcome this through the normal course seasonal build of inventory. We also derived 50.4% of rent and storage revenue from fixed commitment storage contracts in the third quarter, which is 187 basis points higher than the second quarter's level and sets another record for this metric at Americold, while maintaining our low customer churn rate at approximately 3.2% of total warehouse revenues. Rob will go into more shortly, but these key operational metrics illustrate that we continue to perform at high levels for our customers. Second, turning to our priorities around labor management. During the third quarter, we achieved a perm to temp hours ratio of 75:25. This is 300 basis points improvement to our third quarter 2022 permanent labor level and on a sequential basis, roughly flat to the second quarter 2023, due to seasonality when we tend to use more temporary labor in the second half of the year, making the year-over-year metric more relevant. Additionally, we ended the third quarter at an annualized turnover trend approximately 12 percentage points lower compared to prior year. Compared to the end of 2019, a pre-COVID year, we ended September at approximately 11 percentage points higher. This pre-COVID turnover level is an important KPI as we work to improve our services margins. Third, we continue to make progress on our in-process development projects. During the third quarter, we completed our customer dedicated automated project in Russellville, Arkansas that support the large food manufacturer and our multi-tenant automated project in Spearwood, Australia that is anchored by a handful of our top customers. Both facilities went live and we are now in the process of inbounding product as we begin ramping up to stabilization. At this point, 4 out of our 5 automated developments that we outlined at the beginning of the year have completed and launched. With the launching of these 4 facilities this year, Americold is the first and only cold storage company to deliver automated solutions at all 3 key nodes of the supply chain: production advantage, major market distribution, and retail distribution. To have done so in a single year illustrates the enhancements we have made to our platform over the last 24 months to create industry-leading automation capabilities. Additionally, during the quarter, we were excited to break ground on our previously announced expansion project with RSA, our JV partner in Dubai, and we consider this a key step in expanding our relationship with DP World. Lastly, we continue to effectively reprice our warehouse business to offset inflationary pressures in our cost structure to protect margin dollars and to continue to move back to its historical warehouse margin percentages. For the third quarter, rent and storage revenue per economic occupied pallet in our same-store on a constant currency basis increased 3.5% versus the prior year, which was impacted by the reduction of power surcharges in certain markets. Service revenue for throughput pallet increased 6.1%. Moving through the fourth quarter, we will continue to take a surgical approach to our pricing initiatives to continue to drive margin dollars and increase margin percent. At this point, let me comment on our recent common equity raise completed during the third quarter in support of our new growth initiatives. We issued 13.2 million shares off our ATM program at a weighted average price of $31.63 per share for total gross proceeds of 419 million. We utilized all net proceeds to reduce the balance outstanding on our revolver, which decreased our interest expense for the remainder of the year and reduced our current leverage to 5.7x core EBITDA. These proceeds reduced leverage, but will ultimately be used to fund new growth initiatives, while bridging the gap to organic delevering from improved same-store results and stabilizing developments. Turning to growth, today we are excited to announce an approximately $85 million expansion of our Allentown, Pennsylvania facility. As we have said, we are seeing very strong demand from our customer base in key distribution markets within our network and we are well positioned to capitalize on these opportunities. As a reminder, expansion projects are our lowest risk, highest return development projects due to our embedded customer base and the local market knowledge and our ability to utilize our existing operating platform. In addition to expansions, we continue to accelerate the underwriting process around our other two key areas of development focus. First, our customer dedicated build-to-suit developments; and second, our CPKC and DP World collaborations. Additionally, today, we announced the strategic tuck-in acquisition of Safeway Freezers, a temperature controlled company located in Southern New Jersey for a total investment of approximately $37 million. New Jersey is a strategic market for Americold, where we own 15 facilities and we fulfill a variety of customer needs, such as major market distribution, citrus repackaging, retail distribution, protein import and export, and other port services. This acquisition complements our existing portfolio in this market. Before turning to our third quarter results, I am pleased to share that we recently added 2 new executives to our management team. First, Bryan Verbarendse has joined us as our new Chief Operating Officer of North America. Bryan is a supply chain leader with over 30 years of experience in retail and wholesale grocery supply chain at Albertson's. His background and experience are particularly relevant as we look to drive customer service, improve productivity, and increase our services NOI margins. Second, Nathan Harwell has joined us as our new Chief Legal Officer. He joins us from U.S. Xpress Enterprises, one of the nation's largest trucking and logistics services companies and brings deep experience addressing legal, operational and strategic issues across a variety of companies and industries. We are very excited to have Bryan and Nathan on the Americold team. Turning to our third quarter results, we delivered AFFO per share of $0.32, an increase of over 10% versus prior year's quarter. This performance was primarily driven by our global warehouse same-store pool, which generated NOI growth of 5.3% versus prior year on a constant currency basis. Our strong same-store pool results were driven by meaningful economic occupancy growth and pricing initiatives, partially offset by reduced throughput volumes. Reduced throughput was primarily driven by temporary changes to end consumer demand and behaviors due to the challenging economic environment, which has been mentioned by several large food manufacturers and retailers. Some of the factors driving this end consumer behavior includes historically high inflation impacting food prices, reduced SNAP benefits to certain end consumers, the restart of student loan payments, and higher floating interest rates impacting all forms of consumer debt, to name a few. These temporary end consumer behavior changes include less pantry stocking, actively reducing at-home inventories, cooking more meals from scratch and more use of leftovers. We do not believe GLP-1 weight loss drugs had any impact on third quarter results, but let me briefly share initial observations related to the temperature controlled food supply chain based on what we know today. Large food manufacturers and retailers have indicated there has been a little or no volume impact attributable to GLP-1 drugs to date. There also seems to be consensus developing that GLP-1 drugs may impact categories such as soft drinks, snacks, potato chips as an example, and candy. None of which are meaningful temperature controlled categories. The food industry led by food science driven innovation has successfully developed products to meet changing consumer behaviors and diet nutrition for much longer than my years in the industry. To summarize, our global temperature controlled product portfolio is incredibly diverse and includes proteins, agriculture products and prepared packaged foods with each category having excellent nutritional options for a healthy diet today and we are confident that will remain the case in the future. For the third quarter, our same-store economic occupancy increased 345 basis points over the third quarter 2022 to 84%. On a sequential basis, occupancy declined 80 basis points from second quarter, which was slightly less than the 100 basis points to 200 basis points we anticipated due to a seasonal lift. Our same-store throughput volumes declined by approximately 900 basis points versus prior year, primarily driven by the temporary changes to end consumer demand and behaviors due to the challenging economic environment. Despite the 900 basis point drop in throughput volumes, we were able to deliver services margins of 2.8%, which is approximately 30 basis points better than the first half of the year through aggressive variable cost management. We have said the services business has an approximately 50% fixed cost structure, but we can still make meaningful progress on margin improvement by focusing on the variable portion of our cost structure. It's very encouraging to see us improve services margins with throughput volumes down, as it shows that our productivity improvements are beginning to take hold. Heading into the fourth quarter, we expect economic occupancy and throughput volumes to rise sequentially from the third quarter, as our customers ramp up for the normal course holiday season. Not surprisingly, given the economic climate, most large manufacturers and retailers have started promotional activities designed to bring more end consumers into the store to support the holiday season through programs like increased store flyers, couponing and buy one get one free. However, even with aggressive promotional activity we do expect throughput volumes to continue to be a headwind year-over-year given the challenging environment that the end consumer is facing. Turning to full year guidance, as a result of the progress we have made around economic occupancy in our same-store pool, in combination with our ability to manage all aspects of our variable cost structure and the reduction of interest expense due to the paydown of debt, we are raising our full year 2023 AFFO per share guidance up by $0.02 from a midpoint of $1.25 to a midpoint of $1.27 within a revised range of $1.24 to 1.30. Lastly, before I hand it over to Rob, let me comment on our sustainability initiatives, which is a key priority for us here at Americold. I am happy to report we recently received our 2023 GRESB score of 80, which is an improvement of 5 points versus last year's score. Additionally, against our peer set, we also improved our rank to 1st versus 2nd last year. We are very pleased with this outcome and look forward to continued progress in our sustainability journey. With that, I will turn it over to Rob.

Thank you, George. As George mentioned, our company delivered strong results during the third quarter. Economic occupancy at 84% for the same-store pool and another quarter of record setting fixed commitment percentage levels for our total warehouse segment. At quarter end, within our Global Warehouse segment, rent and storage revenue from fixed commitment contracts increased on an absolute dollar basis to $551 million, compared to $379 million at the end of the third quarter of 2022. On a combined pro form basis, we derived 50.4% of rent and storage revenue from fixed commitment storage contracts, which is an approximately 950 basis point improvement over the third quarter 2022. This marks the first quarter in the company's history where more than 50% of total rent and storage revenue has been generated from fixed storage contracts. Since our IPO in 2018, we have added over $350 million in fixed storage revenue, a testament that speaks to both the benefits our customers derive from the structure, along with our best-in-class commercial practices. We're very pleased with this continued progress, in particular, the meaningful progress that has been made this year in recommercializing our European platform, as we transition more of that business to our fixed commitment structure. Turning to pricing, for the third quarter, rent and storage revenue for economic occupied pallet and our same-store on a constant currency basis increased by 3.5% versus the prior year. Please note that during the third quarter, we reduced power surcharges in certain markets, which was a headwind to our increase by approximately 150 basis points to 200 basis points. Service revenue per throughput pallet increased by 6.1%. We remain very focused on our pricing initiatives to ensure that we both offset inflationary pressures and price our business to reflect the value of the service we provide to our customers. We're continuously investing in our facilities, employees, technology and sustainability in order to provide the industry leading service our customers require. We will continue to take a surgical approach to pricing the renewal of existing business along with embedding rate escalation that reflect the current operating environment. We're pricing new business with a forward view of our cost structure and the current market rates and we will also implement our annual general rate increase or GRIs, the majority of which are implemented in January each year. Within our Global Warehouse segment, we had no material changes to the composition of our top 25 customers who account for approximately 48% of our global warehouse revenue on a pro forma basis. Our churn rate continued to remain low at approximately 3.2% of total warehouse revenues consistent with historical churn rates. Given our strong operating metrics, we are continuing to accelerate the underwriting process and evaluating development opportunities across the 3 primary areas of focus that George has mentioned previously. Expansion projects, customer dedicated build to suit developments, and in our CPKC and DP World collaborations. Combined, this macro backdrop along with our strengthened development platform positions us well to capitalize on these potential opportunities. Let me comment on today's announcement. First, we are excited about announcing our plans to build a conventional multi customer expansion project on our Allentown, Pennsylvania site that will be approximately 37,000 pallet positions and approximately 15 million cubic feet. We estimate this total investment to be approximately $85 million and the expansion facility will be anchored by some of our largest food manufacturing customers. We expect to break ground on this expansion in the second quarter of 2024. There are multiple other key markets where demand currently outstrips capacity that we are also looking at developing. In many of these locations, we already own land and are in the final stages of underwriting. Second, during the quarter, we broke ground on our previously announced expansion project with RSA, our JV partner in Dubai. And we consider this a key step in our expanding relationship with DP World. As a reminder, DP World operates over 80 port terminals around the world, many of which lack temperature controlled infrastructure. In addition to the DP World partnership, we are actively underwriting projects tied to our CPKC collaboration. This collaboration will create significant supply chain value to our joint customers by having new Americold facilities is strategically located on CPKC's vast rail network. Over the next 5 years, we expect $500 million to $1 billion of development opportunities combined from these two strategic partnerships. Lastly, our customer dedicated pipeline continues to grow and progressed as customers refocus their efforts on long-term planning having moved past the disruptions from COVID. Our long term relationships with these large customers position us well to secure these build-to-suit opportunities. As for our current in-process developments, as George mentioned, during the third quarter, we completed our customer dedicated automated project in Russellville, Arkansas that supports a large food manufacturer and our multi-tenant automated project in Spearwood, Australia is anchored by a handful of our top customers. Both facilities went live and we are now in the process of inbounding product as we begin ramping to stabilization. Looking forward, we now expect to complete our customer dedicated automated facility in Plainville, Connecticut by the end of fourth quarter and for it to begin inbounding product into the facility by early 2024. This project has a 20-year fixed commitment in place with a leading global grocer. And once we start inbounding, we will begin generating a 100% of the underwritten rent and storage revenue from this project. We look forward to servicing our customer in this facility. Completion of these five facilities represents over 500 million in capital investment being delivered in 2023. Selectively these 5 facilities add approximately 145,000 pallet positions and 46 million cubic feet to the Americold network across all 3 nodes in the supply chain. The delivery of these advanced automated facilities demonstrates the power of our development and automation platform and will create tremendous value for our customers. Lastly, subsequent to quarter end, we completed a strategic tuck-in acquisition of Safeway Freezers in Southern New Jersey for a total investment of approximately $37 million. The Safeway facility as a major market distribution center that consists of approximately 16,800 pallet positions, totaling approximately 6 million cubic feet. Safeway has a very strong customer base, consisting of some new and current Americold customers in the bakery, dairy, potato and juice product lines to service grocery retailers and quick service restaurants. As George mentioned, New Jersey is a very strategic market for Americold, where we fulfill a variety of our customers' needs. And inclusive of this acquisition, Americold now owns 15 facilities totaling approximately 320,000 pallet positions in New Jersey. Now, I'll turn it over to Marc.

Thank you, Rob. Today, I will discuss our net investment activities, our recent capital markets activities, our capital position, and liquidity. I'll then provide an update on our full year guidance. During the third quarter, we completed the previously announced purchase of 1 distribution facility in Brisbane, Australia for a total investment of approximately AUD36 million. Additionally, as George and Rob discussed, subsequent to quarter end, we completed the acquisition of Safeway Freezers for a total investment of approximately $37 million at approximately a 9% net entry NOI yield. We funded these investments through a combination of available cash and our multi-currency revolver. On the disposition front, during the third quarter, we completed the sale of Compreo in Brazil, which resulted in a de minimis amount of net proceeds to Americold. Moving to our balance sheet, during the third quarter, we issued 13.2 million shares of our ATM program at a weighted average price of $31.63 per share at a total gross proceeds of 419 million. We utilized all the net proceeds to reduce the balance outstanding on our revolver, which decreased our interest expense for the remainder of the year. As George mentioned, these proceeds reduce leverage, but will ultimately be used to fund new growth initiatives. This capital raise better positions the balance sheet as we turn to growth. At quarter end, total debt outstanding was $3.2 billion. We have total liquidity of $824 million consisting of cash on hand and revolver availability. Our net debt to pro forma core EBITDA was approximately 5.7x. At this point, we have invested $159 million in our Plainville, Connecticut project in process, and have approximately $32 million remaining to invest on this and other recently completed projects throughout the remainder of the year. Additionally, as George and Rob discussed, we recently announced an expansion of our Allentown, Pennsylvania facility at a total investment of approximately $85 million. We expect to break ground on this expansion in the Q2 of 2024. Please see page 38 of the IR supplemental for additional details on our development projects. As George mentioned, we are tightening our full year 2023 AFFO per share guidance to a new range of $1.24 to $1.30, an increase of $0.02 at the midpoint. Please see Page 40 of the IR supplemental for the key components underpinning this guidance. At this point, I will comment on the primary building blocks to get to AFFO per share and provide a bridge for each as it relates to the full year. Please note the comparisons described represent comparisons to the corresponding prior year results. We are now expecting constant currency revenue growth in the same-store pool for the full year to be in the range of 3% to 5%. Year-to-date it was 5.4%. This implied the fourth quarter to be in the range of negative 4% to positive 3%. Let me provide more detail around the key drivers of this range. For occupancy and throughput volumes. For the full year, we expect economic occupancy to increase by approximately 425 basis points to 475 basis points. Year-to-date economic occupancy increased by 597 basis points. This implies economic occupancy for the fourth quarter to be plus or minus 100 basis points compared to the prior year. We expect to continue benefiting from recent commercialization efforts translating to higher fixed commitments, potentially offset by the impact of manufacturers reacting to the challenging economic environment. For the full year, we now expect a decline in throughput volume of 6% to 7.5%. Year-to-date throughput volumes decreased by 6.7%. This implies throughput volumes decreased during the fourth quarter to be approximately 4% to 10% as end consumer demand continues to slow and basket sizes shrink due to the current economic environment and the factors discussed earlier on the call. For pricing, for the full year we expect constant currency rent and storage revenue for economic occupied pallet growth to be in the range of 5% to 5.5%. Year to date it increased by 6%. This implies growth for the fourth quarter to be approximately 2% to 4%, reflecting our prior year pricing initiatives and the impact of the reduction of power surcharges in certain markets. Also for the full year, we expect constant currency service revenue per throughput pallet growth to be in the range of 6% to 7%. Year to date, it increased 7.2%, this implies growth for the fourth quarter to be approximately 2% to 6%, again reflecting the impact of our prior year pricing initiatives. For the full year, we are now expecting same store constant currency NOI growth to be in the range of 11.5% to 14.5%, which is approximately 850 basis points to 950 basis points higher than the corresponding revenue growth. Year-to-date, same-store constant currency NOI increased by 14.6%. This implies growth for the fourth quarter of 2% to 14%. We are expecting the primary driver of NOI growth to come from rent and storage. Please note, the following guidance metrics are provided on an actual dollar basis, not on a constant currency basis. Turning to the non-same-store pool. For the full year, we expect the non-same-store pool to generate approximately 6 million to 10 million of NOI. Year-to-date, the non-same store pool generated approximately 2 million of NOI. This implies the fourth quarter to in the range of approximately 4 million to 8 million in NOI. During the third quarter, we completed our Spearwood expansion and move the legacy site to the non-same-store pool. Turning to our Managed and Transportation Services NOI. For the full year, we expect these segments combined to generate approximately 46 million to 49 million of NOI. Year-to-date these segments generated approximately 35 million of NOI. This implies the fourth quarter to be in the range of approximately 11 million to 14 million. Turning to our SG&A expense. For the full year, we expect total SG&A to be in the range of 224 million to 228 million inclusive of 23 million to 25 million of stock compensation expense Year-to-date SG&A expense was 169 million inclusive of 18 million of stock comp expense. As a reminder, we exclude stock comp expense from our total SG&A expense to arrive at what we call core SG&A expense, which is what truly impacts AFFO. For the full year, we expect core SG&A to be in the range of 201 million to 203 million. Year-to-date core SG&A was $151,000,000. This is driven by increased cost control efforts given the challenging economic environment, the incremental capitalization of certain costs associated with resources dedicated to the company's in process automated development and the timing of efforts related to our ERP project. Turning to our interest expense. For the full year, we expect interest expense to be approximately 138 million to 141 million. This reflects savings related to the pay down of revolver borrowings following our capital raise during the third quarter and higher interest capitalization related to the delay in the completion of our Plainville development project. Onto our cash tax expense, which is the number that impacts AFFO. For the full year, we expect this expense to be approximately 7 million to 9 million year-to-date it was 6 million. Turning to our maintenance capital expenditures. For the full year, we expect this investment to be approximately 70 million to 80 million. Year-to-date it was 60 million. We have lowered the range for maintenance capital expenditures to measure with a slowdown in throughput activity, as many of our preventative maintenance activities are based on usage. Inclusive of today's development announcements, we expect aggregate starts of 85 million and an additional capital commitment of 5 million to our joint ventures to fund development. Please keep in mind that our guidance does not include the impact of acquisitions, dispositions, or capital markets activity beyond which has been previously announced. Finally, please refer to our IR supplement for details on the additional assumptions embedded in this guidance. Now, let me turn the call back to George for some closing remarks.

Thanks, Marc. We have made significant progress with respect to key operational improvements such as fixed commits exceeding 50% of our global rent and storage revenue, and showing sequential improvement in services margins in the face of throughput volume headwinds. Our latest two automated developments going live give us automated and conventional product offerings at every node in the supply chain. Our capital raise, which improved our balance sheet will provide the fuel for growth in support of our three development initiatives. First, our expansion projects such as Allentown, which was announced today; second, our customer dedicated build-to-suit developments; and third, our CPKC and DP World collaborations. In closing, I'd like to thank the 15,000 Americold associates around the world to their hard work and dedication in servicing our customers every day. It is their efforts that provide the foundation for our future. Thank you again for joining us today and we will now open the call for your questions.

[Operator Instructions] The first question we have is from Samir Khanal of Evercore ISI. Please go ahead.

Yes, good evening everyone. Hey George, throughput is down as we expected and as we had talked about you lowered cost to mitigate the impact. But I guess if throughput remains challenged in 4Q and into the next year, is the strategy to continue to lower cost to mitigate the impact? I know it's variable cost, but I guess is there a limit to how much you can reduce cost there?

There are limits Samir, but when I talk about better variable cost management, we have a ways to go. So, we had what I would call our first quarter where labor productivity really showed up on the P&L with sequential improvement in handling margins with sequential significant sequential declines in throughput. That's the first time we've seen that since we've been trying to manage labor, for the better part of two years now, so I'm very encouraged by that. We can we can do more on the variable labor front, we intend to do more, and we tend to do more, and we should see sequential improvement in margins going forward, even in the face of declining throughputs, although I will say through October we are starting to see sequential increases in throughput, still way off prior year, but sequential increases due to seasonality. So, I'm very positive that we can continue the trend on good variable cost management and it's all about rightsizing labor to meet the work content.

And just as a follow-up, maybe on pricing into next year, how should we think about pricing with inflation moderating, I guess for both storage and rent and warehouse?

Well, we still have our annual GRIs, those happen every year, and we still have renewals. In the case of renewals, they're probably a little bit outsized because, cloud. They're coming up through a very previously high inflationary period. So, maybe you want to go a little deeper, Ralph.

Yes. I would say, you know our main focus is making sure that, renewals are priced up to renewals are priced up to market rates that we're embedding, annual escalation on a go forward basis that reflects the current operating environment. Our GRIs will be implemented like they are every year, which are general rate increases that for the most part go in January. And then when we think about new business, new business is being priced at market rates, which are higher than what we've seen our historical pricing with a lot of our legacy business. So we're confident that pricing will continue to be a lever to help us grow.

The next question we have is from Josh Dennerlein of Bank of America.

George, I wanted to kind of, well, actually no. Thinking about the CapEx guide and how you trimmed it. I guess what's driving that? Is that somewhat a function of like throughput or just like a change in plans? Just kind of help us walk through that.

No. That's a good question Josh. It is directly related to the change in throughput. So part of very good cost -- variable cost management is understanding, the preventative maintenance impact of lower throughput. So let me make an analogy I think everybody can relate to. If you're supposed to change the oil in your car every 5,000 miles, you might drive 5,000 miles in a month, you might drive 5,000 miles in 6 months. If you drive at 6 months, you've deferred that preventative maintenance, 5 months. So, that's essentially what we're doing here in the face of throughput declines, preventative maintenance on our equipment that is used to support throughput, usage of the equipment goes down, and the preventative maintenance spend is extended over time and reduces in this period. So, that's a direct correlation to throughput. If throughput picks up, preventative maintenance would pick up, but so wouldn't earnings from increased handling revenue and profit. So, it's part and parcel of managing variable costs along with labor and other components, but it's directly related to the throughput decline.

And then, since we're on the topic of throughput, you mentioned I think earlier that throughput is going to remain a headwind year-over-year. I guess, is there a way to quantify typically how long these like throughput slowdowns last for and then what happens typically afterwards?

Well, we know that throughput will be down year-over-year in the fourth quarter, there's no question. It will be up sequentially, at least it is in August we believe that's due, I mean, it's up subsequently in October, August. And we know that that seasonality coming into the holidays, we would assume, will continue to see throughput increase sequentially, through Thanksgiving and into Christmas, those are the trends we're seeing, but still well off the prior year. My guess is that in the first half of next year it's very difficult to make meaningful progress on throughput, because the activity in the food industry in the first half of the year. As you know, is seasonally less intense than the second half of the year. My opinion is the first opportunity to see throughput gains would be towards the end of second quarter, when historically we would start to see a pretty significant ramp up for grilling season through the summer, and holidays such as Memorial Day, July 4th, Labor Day, et cetera. So, in my view that's the first opportunity end of the second quarter into the second half of next year to see increased throughput.

The next question we have is from Craig Mailman of Citi.

Marc, I just want to run through guidance real quick because I appreciate the bridge you gave. But by my math, I was looking at the AFFO yield on the equity you raised versus the line that you paid down. And it looks like you're somewhere in the $0.03-ish accretion there plus the $0.05 from G&A. So you're somewhere in the positive $0.08 relative and you guys raised by 2. So does that mean all the core numbers that you're kind of saying that are under a little bit of pressure are kind of negative $0.06 worse than they were as we stood here this time last quarter?

Yes. The first thing is you think about the impact of the interest savings, we estimate the $0.03 savings to be on a full year basis if you look. The impact in year is probably just over $0.01 as it relates to the capital raise. So, that's how to think about that. As you think about the other costs, as George mentioned in his prepared remarks, definitely we are seeing pressure on revenue and contribution from lower throughput. Obviously, I think you heard and you can see in the results that we have made tremendous progress on the cost management side to really mitigate the impact of what we're seeing on lower throughput, I think occupancy stays strong as our really our fourth quarter in a row, I think of record occupancy. So, that's definitely supporting the raise and the guide. So hopefully that helps you bridge those two categories.

That's helpful. Then George, I want to go back to your commentary. You had mentioned that you guys didn't have the 100 basis points to 200 basis point drop in occs that you had expected, because economic was down only 80. But if I look at physical, physical is down 210 basis points. So the 200 plus basis points pickup on fixed commits sequentially seemed to be what really drove it, as the physical kind of looks like it's trending with the year-over-year decline in USDA numbers. So I guess that's number one. And then number two, just kind of curious, you said 9% service margins by the end of next year. Assuming throughput maybe stays weak on the timing you gave, like what kind of ramp do you need to see in the back half of the year to get it there? And is that 9% on, just in the fourth quarter, could you kind of clarify what that 9% means, we should think about it? Yes.

So let me start with the second one first. The 9% is the we said was the run rate we would hit in the second half of the year, so we would achieve that run rate. We wouldn't end the year at 9%, certainly, but we would achieve that in the second half of the year, and we would able to maintain it going forward. The first question sorry, what was the first question?

You guys had thought you'd have 100 basis points to 200 basis points decline was only 80 basis points, but physical was actually down over 200 basis points, which kind of trended with the USDA. So I'm just trying together you guys have said throughput is going to be weak, but are you seeing that even normally seasonal should be picking up now? Are you seeing in the fourth quarter Q4 ahead of Thanksgiving? Or is it still much weaker on a year over year basis?

No, sorry, I forgot that. But, we are seeing sequential improvement in throughput, which should translate into sequential improvement in occupancy. Due to the seasonal lift you just mentioned around Thanksgiving and Christmas, we still believe that will occur and we're starting to see the results materialize in October. The gap between physical and economic did widen, but we believe that the reason why fixed commit their office because people are planning to use that space in the fourth quarter, they've reserved it. We see throughput coming, so, I would expect that gap to narrow in the fourth quarter as large manufacturers are reserving space for their product to support the holiday season.

The next question we have is from Mike Mueller of JPMorgan.

I was wondering, can you give us some color in terms of the economics of the Safeway acquisition and what occupancy levels look like and just economics on the transaction?

Yes. As we said, roughly, we invested $37 million and we expect a net entry NOI yield of approximately 9%. So, as we comment I think Rob commented in his in prepared remarks. The site's located in New Jersey and Southern New Jersey, and it supports our existing infrastructure we have out in that market, which is a very strong market for us.

I think the only thing I would add to that, this is Rob, is just that we'll look to do, with that, acquisition like we do with all acquisitions, implement both our best in class commercial practices and the Americold operating system, which we think should improve that entry yield over time.

The next question we have is from Michael Carroll of RBC.

How abnormal is it for the throughput trend to change that quickly? I mean, if we were in normal times, I'm assuming that this is not a normal time. I mean, how much would we should we expect throughput would vary from quarter-to-quarter?

I don't know that I can say how much we expect quarter-to-quarter, but what I can say is we saw a very, very sharp drop, there's no doubt about it, particularly in the beginning part of the third quarter. It did, as I say, come back pretty strong in the latter part of third quarter and we continue to see throughput improved sequentially through October, but it was a very sharp drop, no question about it, and, I wouldn't consider it normal, but I also wouldn't consider it something that's never happened before. It's just it's not an annual occurrence, let's put it that way. And I think it reflects an economic environment that's fairly unique at the same time.

And then the weakness, was it concentrated in any particular customer, I guess property type like your production facilities, your distribution centers or any part of the region of the company or was it just broad based across the portfolio?

I would say it was broad based across the portfolio, based as we said, we believe on a consumer that has less money walking into the grocery store and is facing higher prices.

The next question we have is from Ki Bin Kim of Truist Securities.

So, we noticed your job postings were down about 40% quarter-over-quarter. I'm guessing that's tied to your variable cost control initiatives. But I just want to tie that to your commentary about this decrease in throughput volume being perhaps temporary, because we've obviously learned that it's kind of hard to retain employees, maybe it's easier to hire than retain, but to make a hiring change like that suggests that maybe this is a little bit longer lasting, so maybe you can provide some more color around it.

Yes, I think it relates to the last question. I mean, the throughput decline was significant and fairly abrupt in the early part of Q3. So you're correct in looking at the job postings and seeing a sharp decline because, what we did was take very quick action when we saw throughput decline very abruptly in the first part of third quarter. That's how you manage variable costs really well. You have to respond very quickly or every day you don't, you're essentially losing money. So I think the fact that you noticed that we cut a lot of job postings tied to a real precipitous drop in throughput, is exactly the correct observation and was very intentional on our part to get labor in line with the work content, which, as I said, dropped pretty dramatically.

And if you look at the cost structure for your -- this is stick with the services business, how much is fixed for a variable? And I know the difficulty in answering that question because it's not linear by a step function, but just trying to get a better sense of how much variable cost can be extracted?

I think a good measure would be 50-50. It's about 50% fixed, 50% variable, what should give you an idea, that if we manage the variable side of this business as well as we can, which is not just labor, it's our maintenance CapEx combined with our labor, if we do that well on 50% of the cost structure, we should be able to expand margins and that's why I say we can still make progress with throughput down. And remember, in occupancy when we were -- about two years ago, we said when occupancy comes back, it comes back at a very high revenue to EBITDA rate because it's incremental. Throughput will do the same thing for that 50% of fixed, right? So we do a really good job on the variable side, and I think this is the first quarter we can point to where we really did. We stay on that track through the first half of next year. In the second half, if throughput improves as many people predict, we should see a really incremental benefit as we absorb more fixed cost and get the benefit of the incremental volume on the variable side. So I'm really encouraged by the progress we've made in this area. We've been doing lot of work as everybody knows around productivity. I would point to this quarter as the first one where productivity really showed up in the results.

The next question we have is from Nick Thillman of Baird.

Please go ahead. Hey, good evening. George, maybe going back to some of your initial comments on SNAP and like lower end consumer. Obviously, you have the COVID SNAP benefits burn off here in February, but now you have the cost of living adjustments kind of kicking in here in October. Is that any way of maybe kicking throughput up or maybe just putting some numbers around just the low end consumer exposure you have in the portfolio?

I think we've seen the low end. I think the low end was the first half, let's say, of the Q3, we've seen sequential improvement since. It really looks like we'll see sequential improvement through the holiday season. That's exactly -- those are the indicators we see. Still off year over year, but not as dramatic, let's say, as a portion of the third quarter. Now the third quarter did recover in the last month and a half or so, I think fairly well, and that's rolling through the rest of the year. But I would say we've seen the bottom, and it was the first half of the third quarter. So hopefully that puts it in a certain context.

Yes, that's helpful. And then maybe, Rob, going through some of your comments on build-to-suit and fixed commitments along development. As we're looking at new starts from here, there's going to be a lot of components of fixed commits associated with them. Are we going to have the traditional J curve? Just, I guess, trying to see what the value creation could be of the stuff that's currently under construction versus maybe some of the takes of new starts in a J curve scenario?

Yes. So I would say our expansion projects, because they're in markets where today demand is outstripping capacity. Our expectation in those expansion projects is that we'll see significant fixed components associated with our expansions. Obviously, in customer dedicated builds, our second main focus from expansions, those will is to be fixed commitments. I think our partnership deals, our CPKC and DP World agreements will probably be is a little bit more mixed and may tend to lean a little bit heavier on the variable side of the equation, just the nature of that business. But categories one and two are expansions and customer build to suits. You should expect to see high fixed content.