Short-term lets: business and regulatory impact assessment

G. Wider economic context

G1. Contribution of tourism and short-term lets to the Scottish economy

143. The tourism sector makes a significant contribution to the Scottish economy. In 2019, there were around 17.3 million overnight visitors to Scotland, spending £5.7 billion^[44], and around 134 million day visits, spending £5.8 billion^[45]. Within this, there were almost 3.5 million international overnight visits, spending around £2.5 billion. The sector generated in the region of £4.1 billion in Gross Value Added in 2018^[46]. Prior to COVID-19, the sector employed 229,000 people, or around 1 in 11 jobs across Scotland, and was a major employer in many rural areas.

144. Data set out in section B1 indicated that, prior to the impact of COVID-19, there had been strong growth in the number of properties registered as self-catering premises on the NDR roll and in listings on online-platform sites. Section B1 also presented data on overall trends in the tourism sector, which did not show growth to the same extent, and set out a number of reasons which could account for this difference.

145. Short-term lets make an important contribution to the tourist economy because they can:

a) offer visitors a unique tourist experience through a host's local knowledge, increasing the attractiveness of Scotland as a place to visit,

b) offer accommodation in places not served by hotels and hostels, for example, and therefore help with dispersal of visitors from "hotspot" areas,

c) offer more affordable accommodation, helping to attract tourists that may have a lower budget, and

d) provide additional capacity to accommodate tourist or other visitor demand in areas with a high demand over a short period of time (for example, to accommodate tourists during the Edinburgh Festival or the Open golf tournament).

146. Sector bodies have undertaken economic impact studies, which are a useful contribution to the evidence base, and which we have discussed with them and their researchers. A study commissioned by Airbnb from BiGGAR Economics in October 2020^[47] estimated that in the region of £0.7 billion gross value added (GVA) and 33,500 full-time equivalent jobs per annum in the tourism sector and wider Scottish economy are supported through spending by Airbnb hosts and guests. Meanwhile, research commissioned by the ASSC from Frontline estimates that spend linked to the self-catering sector supports in the region of £0.7 billion GVA and 24,000 full-time equivalent jobs per annum across the tourism and related sectors in Scotland^[48].

147. The ambit of the studies is not exactly the same: the Airbnb study relates to economic activity booked through their platform, which is the dominant, but not exclusive, online platform in Scotland with nearly 2.5 million visitors to Scotland using accommodation booked through their platform in 2019^[49], while the ASSC study is based on self-catering premises registered on the NDR roll. The hosts and operators covered by the respective studies overlap, since self-catering premises tend to advertise on online platforms. However, the ASSC study excludes home sharing (and therefore B&B) accommodation which is offered on Airbnb.

148. While there will be differences in terms of which hosts are covered, as well as other differences (e.g. estimated spend by guests^[50] and the exact multipliers used), the broad methodology followed by the two studies is similar: survey data on spend by guests is grossed up according to the estimated number of hosts, and then standard GVA and employment multipliers derived from input-output tables are used to estimate the associated economic activity and jobs which are supported by this spend. The studies will therefore have both the strengths and limitations inherent in input-output studies of this kind. The key strength is that the studies will give a good idea of how much activity across the Scottish economy is linked to the spend by short-term let guests. The key limitation is that input-output analysis does not itself answer questions about what the economic and jobs impact would be in response to any change in regulatory or economic conditions. In particular, an input-output study on its own does not answer questions such as:

a) How much would the supply of short-term lets fall in response to the introduction of regulation? The estimated impact of regulations presented by Airbnb and the ASSC in their consultation replies are based instead on separate surveys of hosts, asking what they will do in response to the introduction of regulations (see discussion of this survey evidence in section G3).

b) If short-term lets hosts were to leave the sector in response to regulation, to what extent would this lead to displacement of the activity to other types of accommodation, and to what extent would this represent a net loss in economic activity? For example, Airbnb has argued that if around half of Airbnb hosts leave the sector due to the introduction of regulation, this will automatically lead to half of the associated economic activity and jobs supported being lost^[51], i.e. there is no assessment of whether this could be mitigated by displacement to other accommodation providers^[52].

G2. Impact of COVID-19

149. The trends discussed in section B1 have been interrupted from March 2020 by restrictions imposed following the COVID-19 pandemic and behaviour change curtailing even permitted activities. In 2021, the adverse health impact of COVID-19 has been dramatically reduced by the mass rollout of vaccinations, testing and treatment, and general compliance with public health interventions. However, the economic and behavioural impacts are likely to persist for longer and some changes will not be reversed. So there are now new trends to consider:

a) an increased concern about safety by guests and neighbours, especially around the transmission of COVID-19;

b) additional cleaning costs in between guests as a result of more stringent cleaning protocols due to COVID-19;

c) a likely permanent shift to travelling less for work purposes, now that remote working and videoconferencing have become familiar and, to varying degrees, more popular, and necessary as part of the response to the global climate emergency;

d) lower economic growth relative what was expected prior to COVID-19, which may reduce spend available for leisure travel relative to what it would have been;

e) the potential for some scarring in the tourism sector with some short-term let and other accommodation not returning to the market as part of wider recovery; and

f) behavioural change, and changes in preferences, from guests as experience of COVID-19 affects their risk appetite and priorities, but this may attenuate over time; in some respects a change in preferences may benefit the short-term let sector, as discussed in paragraph 156 below.

150. The tourism economy has been badly affected by COVID-19. In particular, apart from a narrow range of exempted activities, holiday accommodation was closed for a total of 222 days out of 398 days over the period from the start of the first to the end of the second national lockdown. The lockdowns were in place from:

• 24 March 2020 – 3 July 2020^[53] and
• 26 December 2020^[54] – 26 April 2021.

151. While it is currently challenging to estimate the full impact on visitor numbers and spend within Scotland by market, owing to the pause in data collection for both the International Passenger Survey and the Great Britain Tourism Survey in 2020 due to restrictions on data collection arising from COVID-19, other data sources can shed light on recent trends.

152. The Scottish Government's analysis of ONS's Business Insight and Conditions Survey (BICS) statistics, which can be disaggregated to the Accommodation & Food Services sector show that, while COVID-19 exerted significant financial pressure on this sector in Scotland, it is now showing signs of strong recovery, although significant challenges remain^[55]:

• As illustrated in Figure 4, the overall share of businesses categorised as 'currently trading' in the Accommodation & Food Services sector, which had fallen to a low of 32% in mid-March 2021, has recovered to 100% in the most recent wave of the survey (6 September to 19 September 2021), as compared with 99.4% for the economy overall.
• 14.9% of businesses which have not permanently stopped trading in the Accommodation & Food Services sector reported being at severe or moderate risk of insolvency in the period 6 to 19 September 2021, compared with 9.8% for the economy overall.
• Less than one fifth (18.1%) of businesses in the Accommodation & Food Services sector that had not permanently stopped trading reported having no or less than 3 months' cash reserves in the period 6 to 19 September 2021, compared with 21.9% for the economy overall.
• In the period 23 August to 19 September 2021, 45.7% of businesses in the Accommodation & Food Services sector reported experiencing a decrease in turnover compared with what is normally expected for this time of year. The comparable proportion for the economy overall is 29.7%.

153. Note that the Accommodation and Food Services sector has a much larger share of employment in remote rural areas (15%) compared to accessible rural areas (9%) and the rest of Scotland (8%)^[56], so these effects are amplified in more rural areas.

Source: BICS weighted Scotland estimates: data to wave 39

154. Monthly GDP statistics published by the Scottish Government help to summarise some of the above trends, because output is affected not only by the share of businesses which are trading, but also the level of output of those businesses. This data again shows that the impact of COVID-19 on the Accommodation and Food Services sector was significantly greater than for the economy as a whole, especially during the periods of national lockdown, but since then the recovery in this sector has been catching up with the overall economic recovery. As illustrated in Figure 5, relative to its pre-pandemic level in February 2020, output in the Accommodation and Food Services sector reached a low of -84% in May 2020, but by August 2021 it had recovered to 11% below its pre-pandemic level. By way of comparison, total economic activity reached a low of 23% below its February 2020 level in April 2020, and by August 2021 it was 1% below its pre-pandemic level.

Source: Scottish Government, GDP Monthly Estimate: August 2021, seasonally adjusted chained volume measures, gross value added by industry of output.

155. Occupancy data from SuperControl (a provider of online booking technology specialising in self-catering) illustrates that similar trends have been evident in the self-catering subsector. Figure 6, which covers the years 2019 and 2020, shows the impact of the national lockdown in Q2 2020. However, it also shows that, in the second half of 2020, there was a strong recovery in occupancy levels, such that there was in general a higher level of occupancy in the second half of 2020 than there was in the corresponding period in 2019, although the second national lockdown meant that the seasonal spike in demand around Christmas did not happen in 2020.

Source: Frontline report for ASSC, Economic Impact of Self-Catering Sector to the Scottish Economy, Appendix 1 on P24.

156. Figure 7, taken from the Frontline research for the ASSC, suggests that the impact of COVID-19 may have increased the demand for self-catering accommodation relative to other forms of accommodation, with 20% of survey respondents saying that they have now chosen self-catering over going overseas, and 9% saying they have chosen self-catering over staying in a hotel^[57]. This is likely linked to the perceived relative safer conditions in self-catering as a result of reduced exposure to staff and other guests, as well as the greater difficulty of travelling overseas. These impacts may persist for a while, although it remains to be seen if they will be permanent.

Source: Frontline report for ASSC, Economic Impact of Self-Catering Sector to the Scottish Economy, Figure 3.7 on page 19. There were 717 responses to this survey question.

157. The pattern in the SuperOccupancy data is consistent with data from AirDNA (see Figure 8), which also shows that demand for short-term lets recovered strongly in the second half of 2020 following the first national lockdown. The AirDNA data also shows that demand for short-term lets has recovered strongly from the second national lockdown, and in the second half of 2021 it has recovered towards pre-pandemic seasonal levels.

Source: AirDNA data provided to the Scottish Government. AirDNA tracks listings across a number of online platforms, including Airbnb and Vrbo.

158. Different segments of the market may have been affected by COVID-19 in different ways. For example, Figure 9, which shows data from VisitScotland's occupancy survey, suggests that the B&B/guest house sector may have experienced a somewhat greater decline in occupancy than self-catering accommodation, perhaps due to the perception that it is harder to physically distance from other guests/staff in this type of accommodation relative to other forms of short-term let accommodation. However, the latest data suggests that this sector too is recovering strongly back towards more normal occupancy levels.

Source: Scottish Accommodation Occupancy Survey report – B&Bs and Guest Houses

159. While the strong rebound in demand for short-term lets is encouraging in terms of the future prospects for the sector, this is not to understate the financial impact that COVID-19 has had. Firstly, hosts will have foregone significant amounts of revenue during the periods of lockdown, from which they will take some time to recover, even if future occupancy returns to pre-pandemic levels. Secondly, there will be increased costs for some time to come due to the impact of COVID-19, e.g. due to intensified cleaning regimes.

160. As the evidence suggests that the market appears to be recovering towards pre-COVID-19 levels of activity, we are using the 2019 data established in the Scottish Government's 2019 research as a baseline for this BRIA.

G3. Potential impact of this legislation on the tourist economy

161. It is possible that the time and cost involved in applying for a short-term let licence could lead to a reduction in the amount of short-term let accommodation capacity. It could also increase the cost of short-term let accommodation, if these costs are passed on in whole or in part to tourists. There are some uncertainties caused by: the alignment of the fee structure determined by local authorities to the revenue earned by a host; the current condition of the accommodation and what needs to be done to bring it up to standard; changes in visitor preferences, as some may become more confident in using short-term let accommodation, which could act to increase the number of visitors looking to use short-term let accommodation; and the price elasticity of demand^[58] for tourist accommodation. The more inelastic this is, the more short-term let hosts can pass on the costs incurred from the licensing scheme to guests and/or the smaller the reduction in demand for short-term let accommodation from guests.

162. Short-term let control areas could reduce short-term let capacity if significant numbers of planning applications were refused in a control area. It is not possible to quantify the total effect of this as it will depend on whether local authorities implement control areas, the size of these areas, the planning policies that apply and how many short-term lets within a control area already had planning permission.

163. In their response to the 2021 consultation, Airbnb argued that:

The BiGGAR Economics data found that a 50% reduction in Airbnb nights could put 17,000 jobs in Scotland at risk and take almost £1 million a day out of the Scottish economy^[59].

164. As was noted in paragraph 148 above, the input-output study in the BiGGAR study would not in itself be able to predict a 50% reduction. Instead, it was based on the following separate survey undertaken by Airbnb:

We asked hosts in Scotland how the introduction of a registration system, licensing and new taxes on short-term letting would affect how they host. Over half (51 percent) of surveyed hosts in Scotland said they would no longer advertise their space on Airbnb and would not put it on the long-term rental market either. Of that 51 percent:

• 17 percent of hosts surveyed said they would leave their property empty
• 34 percent said they would still live in, but not let their property^[60].

165. The survey asked respondents for their response if a package comprising registration, licensing and new taxes were introduced, rather than just the licensing system itself, but it should also be noted that the survey was undertaken in 2019, before the exact regulatory regime was known, and thus the question did not include the introduction of planning control areas.

166. The ASSC has also undertaken surveys of their members, which they argue similarly indicates that around half of respondents would consider leaving the sector in response to the introduction of a licensing scheme. Key questions from their surveys relating to this issue are extracted in Table 8.

Table 8. ASSC survey questions relating to response by operators to introduction of licensing or registration scheme

Question	Share of respondents	Number of respondents
October 2020 survey
Q42. Would you leave the self-catering sector if this licensing scheme was introduced
Yes	49%	531
No	51%	558
Q43. If the answer is yes, what would you do with the property?
Sell it	27%	253
Leave it empty / use for family & friends	33%	306
Q40. Would you be supportive of a simple online registration scheme, with mandatory health and safety, (akin to Private Landlord Registration) as proposed by the ASSC?[61]
Yes	89%	1,023
No	11%	128
September 2021 survey
Q9. If the Scottish Government introduced licensing as drafted, would you consider removing your property from holiday letting?
Yes	54%	364
No	13%	90
Not sure	32%	214
Q14. How would you feel about your business if 'registered accommodation' was EXEMPT from the proposed licensing legislation (as proposed by the ASSC)?
Very Pessimistic	3%	19
Pessimistic	4%	24
Neutral	18%	117
Optimistic	51%	238
Very Optimistic	25%	170

Sources: ASSC (October 2020), Sectoral Survey into the Impact of the Proposed Licensing Scheme for Short-Term Lets Across Scotland; ASSC (September 2021), Measuring the Unintended Consequences of Short-Term Let Licensing to Self-Catering in Scotland).

167. It is worth noting how different the results are relating to whether the scheme is presented as a licensing or a registration scheme, with 89% of respondents in the October 2020 survey saying that they are supportive of a registration scheme and 93% of respondents in the September 2021 survey saying they would be neutral or optimistic about their business if "registered accommodation" would be exempt as proposed by ASSC.

168. This indicates that the introduction of regulation and adherence to mandatory safety conditions are not in themselves a significant difficulty for the sector; rather the likelihood of hosts actually leaving the sector will depend upon how onerous the provisions, in particular the licence fee, are perceived to be.

169. In terms of how onerous hosts perceived the licensing scheme to be when answering the survey, note that the 2020 ASSC survey included the following two questions:

• Q15. If the cost of a licence were £1000 per property, what would the total cost of fees be to your business?*
• Q16. If an 'oversight' or monitoring fee (akin to Council Tax) was £1000 per property what would the total oversight fee be for your business?*

*this is an indicative cost, based on alcohol licensing fees which range from £800-£2000, since the consultation document does not propose any fee structure.

170. The ASSC also quoted a suggested licence fee in the region of £1,500 to £2,000 per licence in their 2021 consultation response^[62].

171. With respect to what licence fee hosts would have had to pay when answering the questions on registration, Q40 (see Table 8) referred to Private Landlord Registration and also linked to an ASSC paper on licensing versus registration^[63], which also referred in positive terms to the Isle of Man registration scheme. Under the Scottish Government's landlord registration, the current fees are:

• Basic application fee – £67
• Plus additional charge for each property rented out by the landlord – £15
• Late application fee – £133

172. This means that the minimum fee for a landlord with a single property is £82 for registration which lasts three years. With respect to the Isle of Man registration, the ASSC paper quoted a registration fee of £50 for an individual self-catering property, increasing to £100 if registering more than one self-catering property in the same location; if registering property in separate locations, the cost was £50 per property. Registration must be renewed each year, which means that the minimum cost of registration over a three-year period is £150 for a single property, £300 for more than one property in the same location and more if there are properties in multiple locations.

173. As set out in Table 7, the Scottish Government's view is that the average cost of a short-term let licence will be in the range of £214 to £436 for a three-year licence. We are therefore of the view that, for many hosts and operators, the costs will be closer to what survey respondents may have anticipated for a registration scheme than the £1,000 to £2,000 they may have anticipated for a licensing scheme and, in practice, their response will be closer to what they reported for a registration scheme than for a licensing scheme.

174. Note that the ASSC annual membership fee for a host or operator starts at just over £150^[64]; the costs over three years would therefore be at least £450.

175. The ASSC also argued that hosts and operators have very significant concerns about the impact of requirement for a licence to be renewed every three years, arguing that^[65]:

a) taking bookings beyond the period when the licence needs to renewed could be problematic;

b) lenders may be unwilling to lend to a business that may face a licence being refused or revoked;

c) the uncertainty may disincentivise owners to invest in improving their premises; and

d) it may have an impact on insurance cover.

176. The ASSC was particularly concerned about the renewal of a licence being linked to an overprovision policy^[66] since, in their view, it would be difficult for a business owner to quantify the risk that their licence renewal would be refused on these grounds.

177. In October 2021, the Cabinet Secretary for Social Justice, Housing and Local Government announced a number of significant and pragmatic chances to the draft licensing order to respond to concerns raised in the consultation^[67]. These included removing conditions relating to overprovision (on the grounds that the powers given to local authorities to establish control areas are sufficient to implement overprovision policies). Additionally, the requirement for public insurance at a minimum of £5 million was changed to require adequate public insurance; this allows the insurance to be tailored to the particular circumstances, and will make it easier for hosts and operators to use public liability insurance provided by platforms as part of their booking service.

178. As a result of these changes, the Scottish Government's view is that whether the conditions for the renewal of a licence are met will now be substantially within the control of the short-term let host or operator. Provided they ensure that they continue to meet the mandatory safety conditions, their licence would normally be renewed. As a result, owners' incentives to invest in their business should not be diminished. Furthermore, there is no reason why lenders and insurers should be unwilling to extend financial products to such a business, provided that they are satisfied that the business is being run in a safe way, which should be part of their normal due diligence. By way of context, the requirement that private sector landlords and HMO operators renew their registration or licence every three years has not prevented financial products being provided to these sectors.

179. Furthermore, the Scottish Government's view is that design of the licensing system will not impose a disproportionate burden on hosts. Responsible hosts and operators will already be complying with the mandatory safety conditions. A tiered licence fee structure will take account of the size of the premises and the scale of the operation; this is discussed further in section F4. We do not regard it as plausible that half of the short-term lets hosts and operators would leave the sector as a result of the introduction of licensing, or that it would make financial sense for hosts to leave their properties vacant. Furthermore, we think there will be benefits from the licensing scheme to hosts and operators, such as removing unfair competition from operators who are not meeting minimum health and safety standards, which may improve guest confidence in the sector and support demand (see discussion in Section E).

180. Stakeholders have also expressed concern that the licensing system is being introduced while the sector is still being affected by COVID-19. Section G2 presented evidence that occupancy rates have been recovering towards more normal levels, but acknowledged that the negative impact of COVID-19 on the finances of the sector will lag the recovery in occupancy rates. The implementation timetable gives existing hosts and operators time to recover from the impacts of COVID-19 before the deadline for them to submit a licence application (1 April 2023).

181. When considering the effect of the licensing and control area legislation on the wider economy, the effect on areas other than tourism must be considered, such as:

a) Reduced costs on public services:

(i) Less need to contact local authorities or Police Scotland to deal with issues arising from a short-term let, where residents can contact the manager of the premises directly because they know who to contact.

(ii) Reduced numbers and severity of incidents where accommodation meets mandatory safety standards and unsuitable hosts are not granted a licence. This could reduce costs incurred by local authorities, Police Scotland and the Scottish Fire and Rescue Service.

b) Reduced costs from compliance failures. A public register of short-term lets should bring greater transparency and compliance with insurance, tax, mortgage and planning permission conditions.

c) Better functioning housing market:

(i) Local authorities should have greater knowledge about the prevalence and location of short-term lets in their local authority area. This will help them better understand the extent to which short-term lets are, or are not, contributing to local housing pressures.

(ii) Numbers of short-term lets can be managed in areas where there is an insufficient supply of accommodation for local residents. This could reduce housing costs and/or costs to local authorities of temporary accommodation.

(iii) Increased availability of housing from managing numbers of short-term lets could also help local employers recruit workers if they have more choice of accommodation within an appropriate commuting distance. These workers could contribute to the local economy by spending locally, working for local businesses and paying relevant taxes.

d) Improved quality of life for residents: The strength of resident responses expressed in the Scottish Government's research into short-term lets, as well as responses to the consultation process, suggests that there are significant concerns about the effect of short-term lets on local areas in terms of: sense of community, housing supply, safety and local amenities. The licensing scheme and Control Area Regulations will go some way to mitigating these concerns and improving quality of life for residents.

e) Displacement to other tourism accommodation providers: Previous sections of this BRIA have shown that the cost of a licensing scheme should be relatively small in comparison with earnings of a typical host. At the margin, there could be a small number of hosts that exit the sector. But this provides opportunities for other providers of tourist accommodation to meet this demand, helping to maintaining the overall level of employment opportunities in the tourist accommodation sector, as well as wider local economy.

182. These effects will vary across Scotland. Over half of properties in rural Scotland are detached houses (57% in remote rural areas and 50% in accessible rural areas), compared with only 18% of properties in the rest of Scotland. In rural areas over 90% of properties are houses or bungalows, compared with 60% in the rest of Scotland. In contrast, in the rest of Scotland, flats account for 39% of the housing stock, with only 7% of properties in both remote rural and accessible rural areas being flats^[68]. Rural Scotland has a higher proportion of households that are owner occupied (c. 70%) compared with the rest of Scotland (60%)^[69]. Rural areas tend to have less issue with noise and nuisance (not least because of the greater proportion of detached properties) but housing pressures can be acute in tourism hot spot areas; the nearest affordable housing for local people may be at some distance from their work and community. (Note that these housing pressures can affect other tourism businesses, for example hotels needing accommodation for their staff.)

183. The Scottish Government's view is that, in areas where there is a significant reduction in short-term let activity, this will be the result of deliberate and legitimate policy decisions by local authorities using the control area regulations to achieve that aim, rather than the impact of the licensing system. Furthermore, any deliberate reduction in short-term let activity in an area would have compensating economic benefits for that area. For example, a secondary let might be repurposed as a private residential tenancy with a permanent member of the community earning and spending in that community. Visitors who might have used that accommodation are likely to be able to find other accommodation that meets their needs, and it therefore seems unreasonable to assert that the visitor spend would not occur at all. If any reduction in overall economic activity remains, this must be set against the wider social benefits from controlling the level of short-term lets in hotspot areas.