The steep drop in retail demand in Europe, due to measures to contain the pandemic, hurt the first-half (H1) results of retail property owners, particularly in countries where lockdowns were strictest and in place the longest. The wide range of rental income decreases stemmed not only from differences between countries, but also from the type of shopping center, with conveniently located ones faring somewhat better than larger destination malls. Moreover, property valuation declines averaging 2%-5% from January to June 2020 added to pressure in the second half of 2019 linked to a steady increase in online shopping (see "COVID-19 Will Likely Ruin European Retail Property Companies' Efforts To Contain Competition From E-Commerce," published April 1, 2020, on RatingsDirect).

And the pandemic is not over yet. Most of the shopping centers in Europe have reopened but some restrictions are still in place. Retail customer footfall and tenant sales in Europe have since picked up but are not back to last year's levels, and rent collection remains subdued. A resurgence in COVID-19 cases could see a tightening of restrictions. For example, some shopping centers in the U.S. had to close again in July as the COVID-19 situation deteriorated. The number of bankruptcies among European retailers has also increased, and S&P Global Ratings believes negotiations with tenants may lead to lower rental income, casting a cloud over results from July onward. Currently all of our ratings on European retail property companies carry negative outlooks, indicating the possibility of downgrades over the next 12-24 months.

We see a high degree of uncertainty regarding rated retail property companies' results for the full year of 2020. In our revised base case for the sector as of April 1, we forecast a 15%-20% fall in like-for-like net rental income and up to a 10% decrease in property valuations. Yet we are mindful that, for most companies, the H1 results did not reflect the full impact of low rent collection in the second quarter. This is because International Financial Reporting Standards (IFRS) allow unpaid rents, deferred to the second half of 2020, to be recorded as rental income in H1. Whether such deferred rents will actually be collected during the second half of the year remains to be seen. What's more, several companies' H1 reports include a "material valuation uncertainty" clause due to COVID-19. We believe this implies potential additional deterioration of property values in future reporting periods, especially in the absence of sales of shopping centers, which usually provide a benchmark for property valuations.

Revenue Prospects Are Shrouded In Uncertainty

On average, the vacancy rate increased in H1 by 1.2% for the 10 entities we cover in this report (see chart 1), but we've observed some material differences. Mercialys for instance managed to keep vacancies stable at 2.5%, while for a few peers (Steen & Strom, Citycon, and NEPI Rockcastle), the vacancy rate was up by more than 2%. This was due to an increase in bankruptcies among retailers and limited scope to attract new tenants, especially for newly developed shopping centers or extensions opened since the beginning of the year. However, the deterioration of like-for-like net rental income for all European retail property companies we rate was more pronounced and not homogeneous (see chart 2). Four main factors account for this, and we believe they will influence the outcome for several retail property owners over the next two years.

Chart 1

Chart 2

The degree of lockdown restrictions and government support to retailers varies across Europe.  This resulted in notable disparities among retail property companies' H1 results. In particular, we consider this to be the main reason for better operating performance in the Nordics, where lockdown restrictions were lighter, especially in Sweden where shopping centers remained opened. In addition, Nordic governments quickly put measures in place to help retailers pay their rent. As a result, Nordic players' rent collection in the second quarter (based on all rent invoiced) were generally significantly better than for other European entities: Finland-based Citycon collected 88% of rent invoiced, and Norway-based Steen & Strom 67%. Most of the other European companies reported rent collection of only 30%-50% in the first half of the year (see chart 3). We believe retailers' operating performance and, by extension, landlords' rental income will therefore continue to depend to a significant extent on how the COVID-19 situation develops in their markets and government's measures to stimulate the economy or prevent a new wave of cases.

Chart 3

The type of retail property and tenant will influence the shape of the recovery.   Convenient shopping centers have tended to recover much faster than large destination malls, which in some countries stayed closed for longer. Notably, in France, shopping centers larger than 40,000 square meters were not allowed to reopen on May 11 when the lockdown ended, but had to wait until early June. In addition, essential retail stores in small shopping centers seem to attract more people than the experience, fashion, and luxury shopping offers more prevalent in larger malls. As long as the pandemic continues and social-distancing measures prevail, we believe that convenient shopping will remain the preference for customers.

The proportion of stores offering essential goods and services will continue to support rental income.  Having a greater number of such tenants helped shopping center owners' net rental income figures in H1. For instance, essential stores represent about 40% of Mercialys' tenant base, mainly through its exposure to Casino supermarkets. Mercialys reported the lowest drop of like-for-like net rental income in the peer group, namely 0.5%. Similarly, shopping center landlords' relatively low exposure to the fashion and apparel segment helped offset the revenue decline, since this retail segment is one of the most directly affected. This may partly explain the fairly modest 4.1% decline of Citycon's like-for-like net rental income; fashion generally represents only about one-quarter of Citycon's rental income base versus 40%-50% for peers.

We also note that shopping center owners relying heavily on specialty leasing or other types of revenue, like parking fees, were more severely affected than peers, since these lost revenues are not recoverable. This partly explains the 34% drop in like-for-like net rental income for the U.K. portfolio of Unibail-Rodamco-Westfield (URW) in H1, a significantly steeper decline than for the rest of its European portfolio, which is less reliant on these types of revenues.

Different accounting approaches may obscure the full impact of the pandemic for some time.  Under IFRS, companies can report as rental income in H1, unpaid rents in the second quarter that were deferred to the second half of the year. But not all entities took this route for their H1 2020 accounts. This also makes it difficult to compare performance metrics. We may need to wait a few more months to see the full impact of the economic downturn to make a proper comparison. A few companies forgave rents for a couple of months and therefore already recorded those losses in their H1 accounts, while most opted for rent deferrals, followed by negotiations. Discounts granted by landlords against a lease extension could also be streamlined over the duration of the lease, reducing the impact in H1 and potentially 2020. That said, in line with IFRS guidance, some companies also decided to write off a portion of deferred rents already in H1, in cases where they assumed those amounts would not be collected; NEPI Rockcastle, whose properties are mainly in Eastern Europe, is one example. It explains why the drop in NEPI's like-for-like net rental income in H1 2020 was somewhat sharper than for peers, although the impact in future periods should be lower.

Preliminary Post-Lockdown Figures Are Encouraging

Most of the companies in the peer group reported an uptick in retailers' sales and footfall figures in June. Shopping centers performed fairly normally in the first quarter, since the closures started in mid-March. Most were shut in March and April, so sales and footfall were extremely low, but started to recover from May when the levels were at 50%-60% of those in 2019. The improvement was more apparent in June, when shopping centers across Europe were opened for the first full month; sales were about 85%-90% of the June 2019 figures and footfall about 75%-80%, and the recovery continued in July. However, we will continue to monitor how retail sales and footfall develop, since the improvement may be linked to lifting of restrictions allowing customers to make purchases that were postponed during the lockdowns.

Chart 4

Chart 5

Property Values Could Decrease Further

In our base case for the sector, we assume a 10% decline in shopping center valuations this year for most companies. However, overall, we believe valuations could slip lower in the second half of 2020, especially since appraisers remain cautious about their H1 reports. Property valuations declined by 3.4% on average in the peer group as of June 30, 2020, versus Dec. 31, 2019, within a narrow band of 2%-5%.

Chart 6

Citycon reported the lowest decline at 2.1%, but has however already experienced valuation declines over the past few years. On the other hand, URW reported the largest drop of 5.1%. This is because its U.K. and U.S. portfolios were hit more severely (down 13.8% and 6.2% respectively) than its continental European assets (-4.2%), which was close to the European peer average.

We understand that most of the decline in property valuations is directly linked to the impact of COVID-19 on the shopping centers' future cash flows. More specifically, appraisers have adjusted their valuation assumptions to take into account a rent-free period of one-to-three months, some loss of variable rents, an increase in bad debts, and higher vacancy rates, as well as likely reduced re-letting and generally lower rent indexation. The valuation impact from a change in the capitalization rate used to value real estate was in general less significant, as opposed to valuation declines in 2019 that mainly reflected yield expansion, partly offset by positive cash flows. Exchange rates had only a small effect on valuations (notably in the Nordics) that was much lower than the pandemic's effect on cash flows.

All in all, we anticipate a further decline in property value at year end, since we believe appraisers have not yet captured the full impact of the pandemic, as suggested by the inclusion of material uncertainty clauses in valuations as of June 2020.

Tightening Metrics Imply A Shift In Creditworthiness

Our adjusted key ratios for real estate companies, including EBITDA interest coverage, debt to EBITDA, and debt to debt plus equity, deteriorated for the European entities we rate in the first half of 2020, with few exceptions. Some companies still have a buffer relative to our downgrade thresholds, but it is becoming much tighter for most of the others.

Last year, we revised the outlooks on one-third of our ratings on European retail property owners to negative, owing to pressure on ratios from the increase in e-commerce and changes in consumer habits. The COVID-19 pandemic and measures to contain it have accelerated this trend, and ratios deteriorated quicker than we anticipated in 2020. As a result, all our ratings in the sector now carry negative outlooks. In the first half of 2020, we lowered our ratings on two companies, IGD (Immobiliare Grande Distribuzione SIIQ S.p.A.) and URW, which carried negative outlooks in 2019, since their leverage ratios were already elevated for the ratings.

We anticipate that earnings ratios will be materially lower for the full year of 2020, with adjusted debt-to-EBITDA figures likely going beyond our rating thresholds for several companies. What will determine our rating actions however, will be the pace of recovery in 2021 and beyond, since we recognize that the unprecedented impact of the pandemic will distort 2020 figures.

For the time being, we assume a soft recovery in 2021 for the retail property sector, with ratios not returning to 2019 levels for at least two years. The extent to which the pandemic has accelerated structural changes to the retail industry and its consequences for the retail real estate sector are yet to be determined. In our view, it is still too early to assess the medium- to long-term impact on business prospects for retail landlords.

Related Research

• Industry Top Trends Update: EMEA Real Estate, July 16, 2020
• COVID-19 Will Likely Ruin European Retail Property Companies' Efforts To Contain Competition From E-Commerce, April 1, 2020
• Europe's Retail Property Market Is Showing Signs Of Weakening, Sept. 26, 2019