B.2.1. Providers of market accommodation services classified under "Accommodation" (ISIC division 55)

B.2.1.  Providers of market accommodation services classified under “Accommodation” (ISIC division 55)

6.35.                  In this case, the measurements should be broken down according to the categories defined, which are country‑specific. It should be noted that the term “collective accommodation” is no longer used in the United Nations Standard classifications (such as ISIC and CPC). 

6.36.                  For this mode, two main categories of accommodation providers should be established: (a) providers operating within the formal organized economy; and (b) smaller‑scale and occasional providers.

B.2.1.1.              Formal establishments

Modes of observation 

6.37                  As for all productive activities, the observation of establishments classified under “accommodation” must be based on business registers, from which samples of units are selected for surveying. The samples selected for annual, quarterly and monthly observation may differ but should be mutually consistent. 

6.38.                  Countries might have overall business registers (usually held by the NSO) and specific tourism registers (often held by the NTA), whose mutual consistency should be checked. The information provided by business registers on the demography and dynamics of establishments can be of interest in and of itself, especially if it includes data on number of rooms and bed places. 

6.39.                  The coverage of such business registers should be checked, particularly in countries where licensing involves control by the State (or a regional authority) and payment of a specific duty. The updating process should also be verified so as to ensure that sector dynamics are duly reflected. 

6.40.                  In respect of designing the statistical samples, it should be recalled that since tourism activity is often spread throughout a country’s territory, regional dynamics may differ and authorities may be therefore interested in analysing tourism flows by geographical area. This is particularly important for less developed countries, with limited statistical capacity. Such countries tend to concentrate their samples where, in general, most economic and (often industrial) activity is located, whereas tourism may be concentrated elsewhere, e.g., around specific attractions such as sunny beaches, historic sites and beautiful landscapes, and at a distance from centres of traditional economic activity. 

Indicators 

6.41.                  Besides the usual variables (output, value added, employment, consumption and investment), which are common to most economic activities, various other indicators have been developed over the years and are frequently used  to assess the performance of accommodation establishments and of tourism policy: 

• Room occupancy rates (gross or net)
• Bed‑place occupancy rates (gross or net)
• Average number of persons per room
• Average room rate
• Average revenue per room night
• Average revenue per guest night
• (Average) revenue per available room (REVPAR)
• Employees per room
• Average wage per employee
• Revenue per employee 

6.42.                  The use of these indicators requires an understanding of two specific units used in accommodation statistics: 

Rooms

The room is the unit formed by one room or group of rooms constituting an indivisible rental unit in an accommodation establishment. Rooms may be single, double or multiple, depending on whether they are equipped permanently to sleep one, two or several people. Hotels can “sell” or classify double rooms as single rooms, depending on demand. However, a suite whose rooms cannot be rented separately is considered to be just one room. 

Rooms are the units used to measure the capacity of most types of accommodation. Exceptions are campgrounds, for instance, where the unit will be the pitch, and self‑catering apartments, where the measurement unit could include, e.g., a number of bedrooms, a lounge or dining room and a separate bathroom. 

The number of rooms on offer is determined by the number of rooms available for guests during the reference period, including rooms occupied by long‑term guests but excluding those occupied by staff employed in the establishment. This number may be greater or less than the number of existing rooms – usually less, because rooms temporarily unavailable, during low season or because of maintenance, are not counted. On rare occasions, if temporary arrangements are made to accommodate guests in some form of annex, whose rooms are not otherwise available or included in the register, the number of rooms currently on offer may exceed the normal number of existing rooms (in Saudi Arabia, for instance, arrangements might be made with apartment owners if hotels are overbooked during the hajj).

Bed places

The number of bed places in an establishment is determined by the number of persons who can stay overnight in the beds set up in the establishment, regardless of there being any extra beds that may be set up at a customer’s request. The term “bed place” applies to a single bed, double beds being usually counted as two bed places if they are used to accommodate two persons. 

The number of rooms and bed places refers to the capacity in establishments for providing temporary accommodation to visitors.

6.43.                  Based on these units – measured monthly, as is assumed here or, alternatively, on a quarterly, annual or other periodical basis if the country so prefers – the indicators listed above are defined as follows: 

Room occupancy rate. This is an indicator of how many rooms have been sold during the month expressed as a percentage of the number of rooms available (or of the total number of existing rooms) during that same month. If all of an establishment’s rooms have been sold for every night of the month, the room occupancy rate is 100 percent. If only half have been sold, the rate is 50 percent. A “gross” room occupancy rate takes into account all existing rooms declared, whereas a “net” rate takes into account only the rooms on offer. While the net rate is highly useful in evaluating the performance of a given accommodation provider, the gross rate is more useful for macroeconomic study, since accommodation units are usually characterized by the number of existing rooms, as an indicator of size, regardless of the actual number made available on the market. 

Bed‑place occupancy rate. This indicator refers to the number of bed places sold during the month as a percentage either of the number of bed places available or of the total number of existing bed places during that month. In other words, it is the ratio of the actual bed nights sold to the total supply of bed nights (either the number of existing beds or the number of beds on offer). This indicator is similar to the room occupancy rate but provides a better indication of the overall level of an establishment’s occupancy. A bed place occupancy rate that is considerably lower than the room occupancy rate usually indicates that many of the rooms offering two or more bed places are being sold to single occupants[1]. As in the case of the room occupancy rate, the bed‑place occupancy rate can be calculated in gross or net terms, each for different uses, based on the number of bed places actually sold relative to the number of bed places existing (gross) or on offer (net). 

Average number of persons per room. This indicator represents the ration of the total number of guests staying in the establishment to the number of rooms available, aggregated over every night of the reference period. For this indicator, a figure of one means that on average, each of an establishment’s rooms is occupied by one person. A figure of two means that, on average each room is occupied by two persons. Business hotels, which tend to cater more to solo travellers, usually report a lower figure than do holiday hotels, which cater more to families tend to stay. 

Average room rate. This is the average price a traveller pays for a room during the period of reference. It is calculated by dividing the total revenue from room sales for the period by the total number of rooms occupied during the period. It should be net of all taxes separately invoiced. 

Average revenue per room night. This indicator takes into account all of an establishment’s income. Some income is clearly related to rooms: guest expenditures on food, beverages, laundry, telephones, for instance. Other income, however, is not room‑related (but, normally, should still be included in the calculation): e.g., restaurant expenditures by other customers (not occupying rooms), expenditure related to conference centre activities and the leasing of space to shops on the premises. The indicator is calculated by dividing the total revenue from all sales for the period by the total number of rooms occupied, aggregated over every night of the reference period. It is usually expressed in the currency of the country in which the accommodation establishment is located. It might also be useful, for certain purposes, to exclude non‑room‑related revenue from the calculation. 

Average revenue per guest night. Calculation of this indicator is similar to that for average revenue per room night: total revenue from all sales for the period is divided by the total number of beds occupied during the period, aggregated over every night of the reference period. It may be of interest to exclude non‑room‑related revenue from this calculation as well. 

(Average) revenue per available room (REVPAR). This indicator is calculated by dividing total room‑related revenue (core activity) by the total number of room nights available during the period. This indicator is of interest to hotel owners, operators, developers and investors. A similar alternative indicator can be calculated by dividing total revenue (from all activities, namely, core activity plus, e.g., restaurant and spa activity) by the total number of room nights available during the period.  

Employees per room. This is a good indicator of human resource utilization in the sector, and is most useful when calculated from aggregated accommodation data, usually grouped by grade or size of establishment. It is calculated by dividing an establishment’s total number of employees during the period by the number of its total rooms. If the number of employees varies during the period, the average data should be calculated in terms of full‑time equivalent figures. Such calculation also applies to the indicators for average wage per employee and revenue per employee.

Average wage per employee. This is a good indicator of direct employment costs within an accommodation establishment and, when calculated based on aggregate data, can also be useful for a defined group of accommodation establishments. It is calculated by dividing the amount paid for wages and salaries during the period by the number of employees. It should be noted that average labour income per employee might be higher because tips can represent a significant share of employee income (assuming that they have not been included previously within the value of production, value added and remuneration of employees). 

Revenue per employee. This is an indicator useful for comparison with “average wage per employee”, which provides insight as regards earnings in the sector for application within the human resource deployment context. It is calculated by dividing an accommodation establishment’s total revenue for the period by the number of employees working there during that period. 

6.44.                  Usually, such indicators will be compiled from a sample of establishments, stratified according to the different categories of establishment (which might include a geographical dimension), with the results grossed up to the total universe by categories of establishment. There should be alertness during the grossing up process, to the possibility of biases’ being generated in the imputation of non‑responses, especially if the number of units in the sample is small and the rate of response is low. In countries where the total number of such establishments is relatively small, it is recommended that all those establishments be included in the survey (i.e., that a full census such establishments be taken). 

6.45.                  The same indicators as described above can be compiled for enterprises. The caveats are also the same. The geographical dimension is usually lost when the statistical unit is the enterprise. On the other hand, enterprise data are useful for comparing the performance of hotel chains. In addition, indicators on establishments are usually producible on short notice and can be used to satisfy short‑term information needs. Indicators on enterprises (based mostly on annual surveys) reveal more of the structural changes in the industry. 

Table VI.4 

Summary of data on hotels, motels and serviced apartments, 2012 and 2013: example of Australia

Frequency

6.46.              As previously noted, countries might be interested in monitoring the activity of accommodation establishments less frequently: they might prefer an annual or quarterly reference period, for instance, rather than the monthly period assumed above. However, as discussed earlier, tourism is usually a highly seasonal activity, so that data aggregated over a relatively long reference period may not be as useful as data covering a shorter (e.g., monthly) reference period. 

6.47.              For an annual survey, the scope of the information collected by the questionnaire should be broad, including both monetary and non‑monetary data, in order that the economic performance and share of the country’s overall production activity may be properly measured. However, it is recommended that the non‑monetary data be broken down by month or quarter in the annual questionnaire. On the other hand, when it is a question of obtaining more accurate monetary data, an annual accommodation survey should be conducted using enterprises; but non‑monetary data should be derived from establishments (local units).

6.48.              If the survey is conducted on a quarterly or monthly basis, the questionnaire should be much more focused and limited to a few variables, such as those mentioned above, which might also be useful to the establishments themselves, for their own management. Helping establishments develop these indicators and providing them with feedback on the performance of other establishments in their category or region (guided of course by the constraints imposed by confidentiality) might even be a viable means of encouraging their participation. 

6.49.              The use of such indicators on a monthly, quarterly or even annual basis will provide useful information on the performance of the “accommodation for visitors” industry. Additionally, the gross bed place occupancy rate, applied to the total number of existing bed places, provides a measure of the number of overnights by visitors in those accommodation establishments, which might then be compared with the corresponding demand‑side statistics collected through visitor or household surveys (see chapt. III, sects. C.2.2.2, D.1 and D.2). The number of guests and overnights can also be broken down by country of residence and other characteristics of guests.

B.2.1.2.  Informal or small scale providers 

6.50.                  As mentioned above, there are various means of measuring the activity of market accommodation providers that fall beneath the capacity threshold for inclusion of their accommodation activity in the stated coverage, through either a licensing procedure whose scope extends to informal or occasional providers, or an organization representing all or a relevant proportion of providers (e.g., an accommodation industry association). On the other hand, there may be no such organization.  

6.51.                  Where such organizations do exist, it should be possible to obtain their cooperation, at least on an annual basis (or on the basis of specific seasons previously defined), in regard to reporting on the number of such units and setting up simple surveys to gather general figures on occupancy and income. It might be more challenging to ascertain the number of persons staying with such providers, which in some cases may not be of interest to the providers themselves, particularly in the case of apartments or villas. Some of those organizations might be in a position to provide relevant aggregated information collected from their members.

6.52.                  If an effective licensing scheme or organizational system does not exist, or in cases where operators function outside such schemes, it will be extremely difficult to identify their existence and measure their supply, with a solution being possible only through household surveys or population and housing censuses, (see also para. 6.28). If a survey is used then, its coverage needs to be wide enough to enable statistically significant measurements to be obtained; one approach might be to develop a special module for regions where such accommodation units are known to be present in significant numbers. However, such surveys or censuses will cover only domestic visitors and thus will tend to undercount guests in such accommodation units if the guests include visitors who are residents of other countries.