Pandemic will not be the death of cities - Capital Economics
Global Economics

Pandemic will not be the death of cities

Global Economics Update
Written by Neil Shearing

The pandemic – and the associated increase in working from home – may cause a fundamental shift in the way that cities function in future. But this shift will not necessarily trigger a more fundamental economic decline in the world’s largest urban centres.

  • The pandemic – and the associated increase in working from home – may cause a fundamental shift in the way that cities function in future. But this shift will not necessarily trigger a more fundamental economic decline in the world’s largest urban centres.
  • It is easy to see why the pandemic might pose a fundamental threat to the future of cities. For a start, it has revealed that millions of jobs can be done remotely. This raises the possibility that we are facing the “death of the office” and, by extension, city centre working. At the same time, if the shift to working from home is permanent, then people are likely to attach more weight to space when deciding where to live – raising the prospect of a flight from cities, where space is at a premium.
  • A broad and sustained retreat from cities would reverse several centuries of urbanisation and cause substantial economic and social dislocation. It would also have significant investment implications, particularly for property markets. So how seriously should we take the threat?
  • Most economic analysis of the growth and decline of cities tends to emphasise the importance of either local issues (such as connectivity or industrial structure) or big over-arching trends (such as demographics). Macro considerations tend to get less attention but are also important. Two points are worth stressing.
  • First, a country’s capital stock tends to be concentrated in its cities (this includes transport infrastructure, housing, commercial real estate, and digital networks). This creates a barrier to cities shrinking sharply in a very short space of time. Second, cities bring economic benefits in the form of agglomeration and connectivity, which, among other things, spark innovation. As a result, productivity tends to be higher in cities. One study in the US found that productivity in cities was 50% higher than in non-metro areas.
  • There are other non-economic factors to consider when thinking about the future of cities too. For a start, city centres have a much lower carbon footprint than elsewhere. (See Chart 1.)
  • More fundamentally, cities serve as hubs for human interaction. For this reason, they are home to the world’s great cultural, educational, scientific and governmental institutions. Despite the pandemic, humans will remain innately social creatures.
  • These factors help to explain the historic resilience of cities to physical shocks, including war, catastrophic natural disasters such as earthquakes, terrorist attacks and, indeed, pandemics.
  • This is borne out in population data. For example, Warsaw was devastated in the Second World War. Its population fell from around 1.3m in 1939 to just 140,000 by 1945. But it had recovered to over 1m by the end of the 1950s and surpassed its pre-war level by the end of the 1960s. (See Chart 2.) The recovery in the populations of Nagasaki and Hiroshima after the US dropped atomic bombs on the cities was similar.

Chart 1: Average CO2 emissions per capita in the UK by location, 2017 (tonnes)

Chart 2: Population of Warsaw (Millions)

Source: Department for Business, Energy & Industrial Strategy

Sources: American Geographical Society, Statistics Poland

  • Likewise, natural disasters have tended to have a temporary impact on the populations of cities. The 1906 earthquake and subsequent fire in San Francisco killed 3,000 people and left half of the city’s residents homeless, but it did not fundamentally alter the growth trajectory of its population. (See Chart 3.)
  • And it’s worth noting that previous pandemics have not had a lasting impact on the growth or importance of cities. The deadliest pandemic in history was the Black Death that swept through Asia and Europe in the 14th century, killing up to two million people in the process. But while the population of London halved between 1348 and 1350, it had recovered to its pre-pandemic level by 1375. Likewise, the Great Plague of 1665 and the Spanish Flu of 1918 did not have a lasting impact on London’s population (or, in the case of the Spanish Flu, the populations of other US and European cities affected by the pandemic).
  • Instead, when cities have diminished in importance and stature it has normally been for economic reasons. This can be due to a more fundamental economic decline at a national level. For example, having increased from less than 500,000 in 1850 to a peak of over 2 million by 1914 as the power and reach of the Austro-Hungarian empire grew, the population of Vienna then fell following the dissolution of the empire in 1918, reaching 1.5m by the end of the 1970s. (See Chart 4.)
  • The decline in Rome’s population was even more dramatic. It reached nearly 1.5m at the height of the Roman Empire in around 100 A.D. but then fell to less than 100,000 after the empire collapsed. Incredibly, it took until 1950 for Rome’s population to return to the peak recorded at the height of the Empire. (See Chart 5.)
  • More often, a city’s economic decline is driven by the decline of a key sector on which it depends. New York lost 300,000 manufacturing jobs in the early 1970s, presaging a period in which the city’s population stagnated. (See Chart 6.)
  • The city’s population only began to increase again following the rapid growth of its financial sector in the 1990s (and was unaffected by the 9/11 terrorist attacks). The impact of foreign competition and de-industrialisation was even more dramatic in cities in the US “rust belt”, which, unlike New York, failed to develop new industries to replace those in long-term decline. The population of Detroit has shrunk from over 4m in 1970 to less than 3.5m today.

Chart 3: Population of San Francisco (Millions)

Chart 4: Population of Vienna (Millions)

Chart 5: Population of Rome (Millions)

Chart 6: Population of New York City (Millions)

Sources: US Census Bureau, Statistics Austria, ISTAT, United Nations

  • This is not a new phenomenon. Several towns in the UK have experienced a boom and then collapse in populations linked to a change in fortunes for key economic sectors. For example, the population of Glasgow peaked at just over 1m in 1950 but has fallen to less than 600,000 today following the decline of key industries including shipbuilding.
  • The demise of key industries often sets in train a spiral of decline in cities. A shrinking local tax base leads to cuts in services, which then causes a rise in social deprivation and crime, the onset of middle-class flight and a further reduction in the tax base. This can be particularly acute in countries such as the US where there is a significant degree of fiscal de-centralisation.
  • Viewed through this lens, the greatest threat to cities now is not the pandemic itself but rather that a sustained shift to working from home triggers a repeat of the middle-class flight that afflicted many cities, particularly those in the US, in the 1970s. This would have significant spillover to support services – one study by Enrico Moretti at Berkeley found that every job in a “knowledge” industry sustains five others in services, from transport to sandwich shops.
  • However, the nature of the current shock is fundamentally different from those that have previously caused the decline of cities. In the past when cities have shrunk it has usually been driven by fundamental economic forces that are beyond the control of individual firms. For example, industries have entered structural decline or production has been outsourced to lower cost centres, made possible by globalisation. Or, in the case of Vienna and Rome, it has been the result of broader decline at a national level.
  • This time, however, is different. Decisions over post-pandemic working practices and, in particular, how much time is spent working in an office or at home are largely within the gift of firms. And given the productivity benefits that stem from working in close proximity more generally there are good reasons to think that firms (and employees) will want to retain a large degree of office working in future.
  • While it is still early days surveys of firms and workers bear this out. A recent survey of 15,000 Americans by Jose Maria Barrero, Nicholas Bloom and Steven Davis found that 22% of all work days were expected to be supplied from home after the pandemic ends. This is consistent with surveys of firms regarding office space. One recent one by Deloitte suggested that most businesses in London intend to reduce their office space by up to 20% in a post-COVID world.
  • In other words, it’s possible that on any given day there will be 20% fewer workers in city centres than was the case prior to the pandemic. That would represent a significant change to the way that cities operate. But both employers and employees anticipate that the vast majority of work will be done from the office once the pandemic is over.
  • This is not to downplay the macroeconomic effects of such a shift in working patterns. The implications for property markets, in particular, are likely to be substantial. There is likely to be an overhang of office space that will need to be converted to other uses, for example residential or leisure. The same is true of the various retail outlets – from coffee shops to clothes shops – that are sustained by and support office work. (This is something that we have explored in detail in our property services and within our “Future of Property” series.)
  • City centre employment in these sectors is likely to shrink. To put this into perspective, if retail employment were to contract by 20% (i.e. in line with the potential fall in office working) it would equate to a loss of around 100,000 jobs in London. The equivalent figure for New York and Paris would be similar. The nature of these jobs means that the labour market adjustment is likely to fall disproportionately on low-paid workers, who have already suffered as a result of the pandemic. There will also be significant implications for public transport. Capacity on commuter routes in and out of city centres will have to be adjusted to meet lower demand, and capacity in suburban transport networks may need to increase.
  • All of this will amount to significant structural change. But the history of cities is one of constant flux. The key lies in adapting to change. Those that have sufficiently flexible planning policies and labour markets, and are therefore able to adjust quickly and effectively, are more likely to thrive in a post-pandemic world than those with more rigid structures and regulations. But history and economics suggest it would be a mistake to conclude that the pandemic will spell the inevitable decline of cities.

Neil Shearing, Group Chief Economist,