LONDON—COVID-19 put new challenges on STR’s forecasting methodology, not only in the way we forecast demand but also in the way we treat and project supply.
(STR is the parent company of Hotel News Now.)
Our traditional supply forecast is mainly based on our pipeline database, where we apply a set of assumptions around attrition and potential abandonment of properties based on historical trends seen in the specific markets.
However, the lockdown measures in place to contain the virus in most countries worldwide have forced hotels to close for several months or delay reopening and many hotels that remained open suffered severe drops in demand. As a result, we decided to model permanent closures, which, according to historical trends, is a likely result of an event of this scale.
Permanent closures during the Global Financial Crisis
Permanent closures have not been a key trend in STR’s supply in recent years, averaging at 0.3% in each year across all continents. This is because of the growing travel demand, the attractivity of the hotel industry and its developing importance as an asset class within commercial real estate investments.
For this reason, permanent closures did not have much of an impact in our supply models and we had to find other ways to look at this. Given the unprecedented challenges of COVID-19, permanent closures could not be modeled only on trends of recent years. It became clear that the challenges raised by the pandemic needed a stronger benchmark with a similar crisis in terms of impact. The best tool we had looking at the past was identified in the Global Financial Crisis.
In our analysis, we looked at the number of rooms closed each year at both continent and country level from 2007 and calculated the percentage of rooms closed based on the previous year’s supply. We discovered that the financial crisis did not translate into immediate closures in the year after the crisis but had a lagged effect with most closures happening between 2009 and 2013.
Source: STR, © 2020 CoStar Realty Information, Inc.
Globally, there was a 2.3% rate of permanent closures during the five years calculated as percentage of the existing supply in 2009, with a total of more than 370,000 rooms closing. Most closures happened in the Americas, which saw a 3.2% permanent closures rate and almost 200,000 rooms permanently closed. Based on the spread of closures for the Americas in 2009 and 2010, the Americas saw a higher proportion of rooms closed.
While the Americas hold the highest permanent closures rate, the Asia/Pacific region excluding China shows the lowest permanent closures rate over the five-year period with a total of 1.7% of 2009 supply becoming permanently closed. This is of no surprise as this continent was the one suffering the least from the Global Financial Crisis. China holds an even lower rate with 1.2% of closures. Europe and the Middle East saw 2.1% and 3%, respectively, of closures from 2009 to 2013 and both continents registered the majority of closures with a lagged effect from 2010 to 2013. Europe saw over 66% of total closures in these three years while the Middle East/Africa recorded 88% of total closures from 2010 to 2013.
Source: STR, © 2020 CoStar Realty Information, Inc.
At the country level, the United Kingdom had a permanent closures rate of 5.7% between 2009 and 2013 with more than 12,000 rooms closing in 2009 and an additional 22,000 rooms closing during the next four years. Spain only saw a 1.5% of permanent closures over the same period and 66% of total rooms closed—accounting for over 5,700 rooms—happened in the last two years of that period. The United States saw a 2.9% permanent closures rate spread almost equally across the five-year period with an average of 27,000 rooms closed in each year.
COVID-19
Despite being an insightful starting point to further understand permanent closures, the Global Financial Crisis benchmark raised some differences and challenges that could not be applied to COVID-19 closures, given the different natures of these events. The Global Financial Crisis’ lagged effect is linked with the economic impact of this crisis.
On the contrary, COVID-19 had an immediate impact on the tourism industry. On 11 March, the World Health Organization declared COVID-19 as a pandemic, which made traveling impossible and forced over 3 million rooms worldwide to close. This is an unprecedented number of temporary closures, and eight months later more than 700,000 rooms worldwide are still closed, accounting for almost 4% of total rooms reported in STR database.
To navigate the issue around the lagged and varied impact of permanent closures as well as COVID-19 exceeding the Global Financial Crisis in terms of impact on the travel industry, we introduced a new methodology comparing RevPAR performance during this COVID-19 period to the worst period seen during the Global Financial Crisis. This was then applied to the permanent closures observed during the Global Financial Crisis to calculate the new rate of permanent closures at market level. For example, if RevPAR was three times worse during COVID-19 than Global Financial Crisis and closures were 1%, then permanent closures applied would be 3%.
Closures are assumed to take place over a three-year period starting in April 2020, compared to five years during the Global Financial Crisis with the assumption that COVID-19 will have more of a significant impact in the short to medium term. The supply was adjusted from April 2020 onwards with the view that a proportion of properties that are temporarily closed from then onwards will never reopen, at which point they will be removed in the data back to the first calendar month they were closed in STR’s systems.
In the case of some markets, the ratio of RevPAR performance comparing COVID-19 to the Global Financial Crisis gives a very high closure rate due to the extreme declines in performance during COVID-19, so to navigate this, the estimated permanent closure rate has been capped at the region level.
Half of these closures are assumed to occur in the first year while the remainder is spread across the following two years. Hotels will be affected more in the short term as lockdown measures and travel restrictions cut off demand, increasing the risk of permanent closures. The threat also remains in the medium to long term as demand is not expected to return to 2019 levels in some cases until 2023 or 2024, which is why 25% of permanent closures have been allocated to each of the following two years.
For Tokyo and Dubai, which in 2021 are hosting the Summer Olympics and EXPO, respectively, the phasing of these closures has been adjusted to expect a lower weighting of closures in the first year than in the following two years. Given the impact COVID-19 has had on events and travels, less demand is expected to flow into these markets when compared to pre-COVID-19, but the expectation is that performance will still see an uplift, meaning this demand may slightly reduce the threat of permanent closures in Tokyo and Dubai.
Source: STR, © 2020 CoStar Realty Information, Inc.
The chart above shows a selection of key forecast markets globally with the estimated permanent closure rate for three years. There is a clear divide between the Europe, Middle East and Africa region and the Asia/Pacific, with Asia/Pacific markets expected to see fewer closures.
In terms of hotel performance, these markets have not been as impacted due to COVID-19 being under control in markets such as Beijing and Auckland. In Singapore, quarantine demand has kept occupancy levels high during the pandemic. Additionally, these markets all observed lower rates of permanent closures during the Global Financial Crisis.
Meanwhile, European markets have struggled to contain the virus with numerous waves of COVID-19 hitting the continent, and coupled with higher rates of hotel closures during Global Financial Crisis, this pushes up the estimated closure rate in these markets, meaning a higher threat of hotels closing or not reopening following temporary closure.
Performance impact
An increase in the level of hotel closures means less supply available in the market. In the long term, this translates to higher occupancies with fewer competing hotels in the market. This can also mean less pressure on ADR as it slightly reduces the need for hotels to discount in order to compete on rates. In London, for example, this new methodology has reduced supply by 2% to 3% each year from 2022 to 2024, leading to an increase in occupancy in both 2023 and 2024.
Conclusion
The Global Financial Crisis is the best benchmark available for predicting permanent closures, but it isn’t perfect. There are large disparities in performance between what happened during the Global Financial Crisis and the fallout from COVID-19, with most markets now experiencing some of the lowest levels of hotel performance ever observed as government-imposed lockdowns and travel restrictions decimate the industry. Estimated COVID-19 closure rates are possible by comparing the performance between the two crises and applying it to Global Financial Crisis closure rate.
The lagged effect seen in closures during the earlier crisis brings challenges in using the Global Financial Crisis as a benchmark. Aggregating closures to a 2009-2013 rate made it possible to navigate that, the assumption for COVID-19 is that the impact will be significant in the short-term with 50% of estimated closures across the three years expected to take place in the first year.
Christian Suarez is a junior analyst with STR’s Research & Analysis team, based in London. Barbara Fraccascia is a forecast and research analyst with STR’s Research & Analysis team, based in London.
This article represents an interpretation of data collected by STR, parent company of HNN. Please feel free to comment or contact an editor with any questions or concerns.