In this paper, we emphasize how the preservation of cultural assets contributes to improving local economic conditions and generating for the society. As a case study, we examine, analyze, and quantify the economic and social value of the Colosseum.
We analyzed and computed the main economic contribution of the Colosseum related to its direct use, an estimate of its indirect use value. In addition to this, we explored and provided an estimate of the non-use value related to the non-commercial aspects of the Colosseum.
Considering both its direct operation and induced tourism, we estimated that the direct use of the Colosseum generates about 1.4 EUR/bn a year in value added to the Italian economy (GDP contribution). In addition to the direct use value, we developed a hedonic pricing model to estimate the Colosseum’s indirect use value for residents is about 0.4 EUR/bn. Using a contingent valuation method, we estimated the total existence value of the Colosseum at an asset value of 76 EUR/bn.
We propose a broader economic analysis able to quantify the material and immaterial values related to an iconic cultural heritage asset estimating different dimensions of its value. The estimated value cover and quantify a crucial cultural heritage aspect and it aims to support the definition of policies regarding the sustainable use of cultural heritage itself and its constituent assets by both local and non-local tourists, as well as to facilitate conservation and restoration initiatives. Therefore, this paper has the potential to provide valuable analysis and estimation that can be integrated into the formulation of more holistic cultural heritage management strategies.
1. Introduction
The Flavian Amphitheater, known as the Colosseum, is the most iconic symbol of Rome and one of the most famous monuments of Ancient Rome. It is also one of Italy’s main tourist attractions. In 2019, more than 7 million people from around the world visited the Colosseum.
This paper examines, analyzes, and quantifies the economic and social value of the Colosseum, focusing specifically on its impact on the Italian economy and society.
The economic value of a cultural heritage site like the Colosseum represents the total well-being it generates for society. This includes both tangible financial benefits and intangible advantages. In fact, the Colosseum’s value extends far beyond its direct economic contributions, encompassing broader societal gains.
As Rome’s primary tourist attraction, the Colosseum generates direct revenue from tourists who visit it and stimulates further tourism by attracting visitors to Rome and Italy who might not otherwise have come or extended their stay. The tourism spending linked to the Colosseum creates direct effects resulting from business activities related to the production of goods and services in the tourism sector. These goods and services are purchased by tourism companies from their suppliers, forming the tourism supply chain. Induced effects reflect the increased household-to-business activity driven by the personal income generated through direct and indirect effects. These effects are primarily related to the consumption patterns of businesses and their employees that have benefited from initial expenditures in the tourism sector. The total value added for the economy generated by the visitors’ expenditure that can be attributed to the Colosseum represents its direct use value. In fact, the direct use of the Colosseum mainly consists of allowing people to visit it. Considering both its direct operation and induced tourism, we estimate that the Colosseum generates about 1,390 EUR/mln a year in value added to the Italian economy.
In addition to this value, another element is given by the value derived from the indirect use of the Colosseum, i.e. by securing some benefits from it. For example, the Colosseum may provide welfare to people just passing nearby and enjoying the scenery without spending money, or in the form of “housing comfort” as people derive welfare from living close to it. In the former example, the indirect use of the Colosseum is not captured by an economic or financial transaction. However, market transactions related to the housing comfort can be observed, although indirectly, in the housing market. The sum of direct and indirect use value lead to what we can refer to as the actual use value of the Colosseum. We estimate that the indirect use value of the Colosseum, in terms of housing comfort for residents, is about 406 EUR/mln. Note that, unlike the direct use value, this is not an annual figure.
Beyond the actual use value, one must consider the non-use value, which arises when an individual is willing to pay for the Colosseum even if they do not make direct use of it, do not benefit indirectly from it, or do not plan any future use for themselves or others. Based on an ad-hoc survey, we found that about 90% of Italian residents believe the Colosseum is an iconic Italian landmark representing the most important cultural attraction in Italy, and that it must be preserved under any circumstances. We also found that Italian respondents are willing to pay a significant amount of money to preserve the Colosseum. Specifically, our estimates suggest that the Colosseum’s non-use value, representing its social value, is 2,936 EUR/mln per year.
The remainder of this paper is organized as follows. In Section 2, we provide a brief review of the literature on economic analysis of cultural or natural heritage assets. Section 3 outlines the methodologies used to evaluate the Colosseum’s economic contribution in terms of direct, indirect, and induced use value. We also assess its non-use value and existence value. In Section 4, we summarize the results of the analyses from Section 3. Section 5 presents the conclusions of the study, and Section 7 discusses the practical and theoretical implications.
2. Review of the literature on economic analysis of historical and cultural assets
The idea of putting a price tag on cultural heritage, mainly aimed at proving that conservation does not only have costs but also benefits, is not new. Indeed, several economic valuation studies have been conducted worldwide. Bowitz and Ibenholt (2009) address some methodological questions regarding economic impact studies of investments in cultural heritage projects. Specifically, different types of direct and indirect impacts are discussed, particularly how these can be calculated. In a study of the Norwegian town of Røros, the authors find that tourism related to the cultural heritage in the region contributes approximately 7% to overall employment and income. Panzera (2022) describes how cultural heritage contributes to enhancing local economic conditions.
In 2013, the European Expert Network on Culture published a study aimed at highlighting the potential of the cultural sector for social and economic development by analyzing which models are used to evaluate cultural heritage. The majority of studies reviewed use concrete case studies to demonstrate the impact of specific heritage sites.
Regarding the estimation of indirect use values, the literature shows that the application of hedonic pricing is well established for the valuation of assets in the real estate market, and that this method has recently been borrowed for the market for cultural assets. Belniak and Wieczorek (2017) apply hedonic pricing to evaluate a set of flats in Lublin, while Lazrak et al. (2014) provide one of the first applications using a spatial autoregressive model to investigate the impact of cultural heritage on the value of real estate in cities. In the first case, the results suggest that the values estimated through the application of this model may be substantially different from the observed transaction prices. In the second paper, the effect of cultural heritage is measured by a specific heritage variable, given by the number of listed buildings within a 50-m radius. The main findings are that structural characteristics have a positive and significant effect, while spatial characteristics are also approximately equal to 21% compared to non-listed heritage.
Regarding the estimation of non-use value, Bateman et al. (2002) provide an in-depth analysis of the application of stated preferences techniques to economic valuation. In particular, they focus on all the steps needed to implement a stated preference study for both contingent valuation and choice modelling, even though they differ in terms of questionnaire design and data analysis.
Several papers estimate the non-use value with regard to specific cultural heritage assets: Beltran and Rojas (1996) apply the Contingent Valuation Model (CV) to estimate people’s willingness to pay (WTP) for the consumption and preservation of three archaeological sites in Mexico; Hansen (1997) estimates the aggregated WTP for the Royal Theatre using a sample of 1,843 Danes; Salazar and Marques (2005) apply CV to estimate the social benefits stemming from the restoration of an old Arab tower in the Valencia region of Spain. In the first paper, the results show that the WTP for visitors who paid an entrance fee is significantly higher than the WTP estimated for non-paying visitors, and that there is a significant dispersion in the WTP expressed by the respondents. Thus, a price discrimination strategy is recommended to address the issue of the decline in resources provided by the government to the cultural heritage sector. Hansen’s paper shows two main findings: the estimated aggregated WTP is not lower than the amount received by the theatre in public subsidies; although the majority of the Danish population never visits the theatre, they are willing to pay an option price for the possibility of visiting it in the future. Thus, it can be affirmed that the theatre also has a non-use value. In the last paper, the main finding is that people were willing to pay much more than the per capita expenditure devoted to the protection of cultural heritage in the Valencia region. In addition to these papers, we can also cite Choi et al. (2010), who employ choice modelling to study the Old Parliament House (OPH) in Canberra, and Diafas et al. (2017), who conduct a choice experiment (CE) to estimate the economic value of changes in ecosystem services that impact the welfare of rural communities near a rainforest in Kenya, measuring the indirect use value of the forest and its bequest value.
Ruijgrok (2006) shows that the economic benefits of conserving the most threatened types of cultural heritage surpass the costs, estimating three different benefits: housing comfort value, recreational value, and bequest value. The first is calculated through the application of the hedonic pricing method, while the second and third are calculated using the Contingent Value Method (CVM).
Economic evaluations of cultural assets are often criticized for their reliance on surveys. Instead, a balanced approach that integrates the perspectives of all stakeholders is recommended, given the complexity and economic value of cultural heritage.
Among the various economic evaluation studies conducted worldwide, several articles and books describe the approaches used in estimating the economic impacts of cultural heritage. Although there is potential, evaluation studies on cultural heritage are scarce. Analyzing the literature, a strong geographical concentration of studies emerges. Most of these come from Europe (55%), with the U.K. producing the largest quantity (29%). About 23% of the studies were conducted in the United States and Canada. It is difficult to make comparisons between existing studies: the nature of the assets analyzed, the types of benefits assessed, and the activities involved differ, so the results and conclusions vary from study to study. However, some studies, despite having different conclusions, are consistent in their results.
In this paper, we provide an analysis of the broad value of a cultural heritage site like the Colosseum, which should be viewed as a distinctive and scarcely comparable asset compared to other cultural heritage sites analyzed in the literature, given its societal positioning, reminiscence power, and historical origin. We look at different aspects, such as its main economic contribution related to its direct use, its indirect use value, and its non-use value, including non-market aspects and capturing its existence value.
3. Methodology
3.1 The economic contribution
The Colosseum may be considered a factor of production, a fixed capital whose main output, at the end of each year, is represented by the returns from tourism.
As a tourist attraction, the direct use of the Colosseum mainly consists of allowing people to visit it. Thus, the first element of the economic contribution of the Colosseum is given by the direct cash flows produced from its specific exploitation. These revenues are derived from the entrance fees tourists pay to enter the Colosseum and from additional services related to the visit, such as guided tours, the purchase of audio guides, ticket pre-sale rights, and royalties on gadgets sold at the bookshops. These direct effects from the use and operation of the Colosseum create additional activity in the local economy, generating indirect and induced effects. Indirect effects concern intermediate consumption for the production of goods and services in the tourism sector, while induced effects concern expenditure by employees from wages paid by companies in direct contact with tourists, as well as the consumption of companies that have benefited directly or indirectly from initial expenditure in the tourism sector.
In addition to the direct tourism spending generated by its operation, the Colosseum contributes to tourism in a broader way by attracting tourists to Rome and Italy who would not otherwise visit, or at least would not stay for the same length of time. Thus, this contribution is another important component of the total economic contribution generated by the Colosseum.
The economic contribution from the direct operation of the Colosseum is measured in terms of its contribution to value added to the Italian economy. The value added within the tourism sector generated by visits to the Colosseum represents a measure of its direct effects, which create additional activity in the local economy, generating indirect and induced effects. The indirect effects are the results of business-to-business transactions indirectly caused by the direct effects. The indirect effect is a measure of this increase in business-to-business activity (not including the initial round of spending, which is included in the direct effects). Furthermore, induced effects are the results of increased personal income caused by the direct and indirect effects. The induced effect is a measure of this increase in household-to-business activity. The total economic contribution of the Colosseum includes both its direct, indirect, and induced contributions to value added.
The methodology for evaluating the indirect and induced economic impacts generated from the direct effects is carried out in an input-output (IO) framework, relying on the Tourism Satellite Accounts (TSA), which provide the most accurate and reliable measurement of the role of tourism in an economy. Alongside the TSA-based analysis, it is possible to evaluate the expected impacts of tourism by calculating its multiplier effects (Table 1). The TSA is based on the principles of National Accounts, an integrated statistical framework that measures a country’s national output from each sector’s contribution to economic activity. The TSA is based on input/output tables, which measure the activity of producers and purchasers of goods and services across the spectrum of economic sectors.
3.1.1 Data
The main data source for estimating the Colosseum’s economic contribution from its direct operation is its 2019 financial report, maintained by the Italian Ministry of Cultural Heritage and Activities. Our analysis in this report refers to the year 2019, because at the time of the study, this was the most recent year for which complete data on all the components of the value used in our analysis were consistently available. It should also be noted that focusing our analysis on 2019 has the advantage of avoiding possible distortions due to the pandemic that spread since 2020.
In this analysis, we use data from the Italian TSA maintained by the Italian National Institute of Statistics (ISTAT), according to international guidelines and standards. Although indirect and induced effects cannot be immediately calculated from the provided data, it provides the basis for measuring the direct and indirect effects of tourism, using intermediate consumption and compensation derived from tourism output. The TSA thus spans both tourism consumption and tourism output.
3.2 Induced tourism contribution
Since the Colosseum is one of the major tourist attractions in Rome (and in Italy), it is difficult to separate its impact from that of other attractions. Thus, our main approach to estimate the contribution of the Colosseum to broader tourism expenditure in Rome identifies only a proportion of all tourists visiting the Colosseum each year as visitors who would not otherwise visit Rome, or at least would not stay for the same length of time. We are implicitly recognizing that many tourists are attracted to Rome by a variety of attractions, while only some of them, and consequently their tourism expenditure, are directly attributable to the Colosseum.
3.2.1 Data collection and analysis
The approach used to estimate tourism expenditure involves first estimating the share of visitors to the Colosseum who would not otherwise visit Rome, and subsequently estimating the average tourism expenditure of international and domestic tourists who take a cultural vacation in Rome. Our estimate of the broader tourism expenditure that can be attributed to the Colosseum is thus given by a weighted average of tourism expenditure, with weights equal to the number of domestic and international visitors to the Colosseum who would not otherwise visit Rome.
In order to estimate the share of visitors to the Colosseum who would not otherwise visit Rome, or at least would not stay for the same length of time, we used responses to an ad-hoc survey submitted to a sample of Italian residents. Respondents were also asked whether the Colosseum was the main factor in their decision to visit Rome. The percentage of respondents reporting that the Colosseum was the main factor in their decision to visit Rome is about 52%. However, since we suspect that this percentage could possibly overestimate the true share of visitors—given that Rome has many other attractions—we adopted a conservative approach. We estimated the share of Colosseum visitors who would not otherwise visit Rome as the percentage of respondents reporting that the Colosseum was the main factor in their decision to visit Rome, and who also ranked the Colosseum as the most important cultural tourist attraction in Italy. This percentage is about 28%.
Then, in order to estimate the average tourism expenditure for tourists who visit Rome for a cultural vacation, we analyzed the profile and level of expenditure of both domestic and international visitors on different categories of expense, namely local transportation, accommodation, food and beverage, purchases in stores, and cultural and entertainment services. Data used to estimate tourism expenditure related to international and domestic visitors comes from the “International Tourism in Italy” survey, conducted annually by the Bank of Italy, and the “Trips and Holidays” survey, conducted by the Italian National Institute of Statistics (ISTAT). Using data from these surveys, we identified a representative sample of foreign and domestic visitors who visited the city of Rome in 2019 and who reported a “cultural vacation” as the main reason for their journey. After selecting these specific visitors, we estimated their average per-person expenditure.
Table 2 shows the results of a regression model for the relationship between the length of stay (number of nights) and total tourism expenditure of both domestic and international visitors, as well as visitor characteristics. In addition to total per-person expenditure, we also estimate expenditures separately for the five main categories of tourism expenditure: accommodation, food and beverage, internal or local transportation, shopping, and entertainment.
Estimated coefficients of the regression model for average length of stay and average expenditure (EUR) per person per day of domestic and international visitors
| Domestic visitors | International visitors | |||
|---|---|---|---|---|
| Number of nights | Total expenditure | Number of nights | Total expenditure | |
| January | −3.780*** | 67.240 | −0.850*** | −24.250** |
| February | −0.390 | 61.770 | −0.980*** | −22.870** |
| March | −3.070** | 74.890* | −0.850*** | −20.710** |
| April | −3.580** | 141.530*** | −0.610*** | −1.320 |
| May | −2.850** | 98.660** | −0.310 | 11.400 |
| June | −3.020** | 37.190 | −0.690*** | 9.560 |
| July | −1.510 | 27.710 | −0.210 | 2.230 |
| September | −2.180* | 11.700 | −0.220 | −14.210* |
| October | −3.220* | 22.770 | 0.290 | −8.640 |
| November | −1.820 | 84.530 | −1.180*** | −11.460 |
| December | −3.530** | 144.760*** | −0.480* | −14.630 |
| Manager | −0.480 | −18.300 | 0.220 | 23.160*** |
| Industrial worker | −1.230 | 36.280 | −0.960*** | −83.200*** |
| Self-employed | 2.970* | −11.860 | −0.180 | −17.380** |
| Student | −0.860 | −22.470 | 0.220 | −52.000*** |
| Retired | 0.790 | 90.020** | −0.560 | −101.600*** |
| Other | −0.370 | 14.990 | −0.630*** | −19.550*** |
| Age 15–24 | 0.030 | 37.100 | −1.500*** | −158.700*** |
| Age 25–34 | 1.280 | 82.780*** | −1.340*** | −134.470*** |
| Age 35–44 | 0.220 | −12.060 | −0.970*** | −63.990*** |
| Age 65+ | −0.680 | −58.700 | 0.540 | −5.700 |
| Constant | 5.350*** | 25.620 | 7.200*** | 366.080*** |
| Obs. | 810 | 810 | 9,176 | 9,176 |
| Domestic visitors | International visitors | |||
|---|---|---|---|---|
| Number of nights | Total expenditure | Number of nights | Total expenditure | |
| January | −3.780*** | 67.240 | −0.850*** | −24.250** |
| February | −0.390 | 61.770 | −0.980*** | −22.870** |
| March | −3.070** | 74.890* | −0.850*** | −20.710** |
| April | −3.580** | 141.530*** | −0.610*** | −1.320 |
| May | −2.850** | 98.660** | −0.310 | 11.400 |
| June | −3.020** | 37.190 | −0.690*** | 9.560 |
| July | −1.510 | 27.710 | −0.210 | 2.230 |
| September | −2.180* | 11.700 | −0.220 | −14.210* |
| October | −3.220* | 22.770 | 0.290 | −8.640 |
| November | −1.820 | 84.530 | −1.180*** | −11.460 |
| December | −3.530** | 144.760*** | −0.480* | −14.630 |
| Manager | −0.480 | −18.300 | 0.220 | 23.160*** |
| Industrial worker | −1.230 | 36.280 | −0.960*** | −83.200*** |
| Self-employed | 2.970* | −11.860 | −0.180 | −17.380** |
| Student | −0.860 | −22.470 | 0.220 | −52.000*** |
| Retired | 0.790 | 90.020** | −0.560 | −101.600*** |
| Other | −0.370 | 14.990 | −0.630*** | −19.550*** |
| Age 15–24 | 0.030 | 37.100 | −1.500*** | −158.700*** |
| Age 25–34 | 1.280 | 82.780*** | −1.340*** | −134.470*** |
| Age 35–44 | 0.220 | −12.060 | −0.970*** | −63.990*** |
| Age 65+ | −0.680 | −58.700 | 0.540 | −5.700 |
| Constant | 5.350*** | 25.620 | 7.200*** | 366.080*** |
| Obs. | 810 | 810 | 9,176 | 9,176 |
Note(s): Coefficients: * significant at 10%, ** significant at 5%, *** significant at 1%. The constant refers to the average expenditure of a representative tourist, who is an employee between 45 and 64 years old, visiting Rome in August 2019. All regressors are binary variables (dummy) and their coefficients can be interpreted as the difference in average expenditure between the category they represent and the representative tourist
Source(s): Table created by authors
Table 3 shows the results of a regression model for the relationship between tourism expenditure on different categories and visitors characteristics. Since for domestic visitors we do not have detailed expenditure data distinguished by category, we estimate the share of total expenditure for each category using data from TSA. Figure 1 shows the share of total tourism expenditure for domestic and international visitors by category.
The two side-by-side pie charts show domestic and international visitor spending across five expenditure categories. The caption lists the categories as: “Local transportation”, “Accommodation”, “Food”, “Shopping”, and “Entertainment”. Data for each pie chart is provided in a clockwise sense. Domestic visitors (clockwise): Local transportation: 36.3 percent. Accommodation: 29.6 percent. Food: 20.7 percent. Shopping: 12.4 percent. Entertainment: 1.1 percent. International visitors (clockwise): Local transportation: 9.6 percent. Accommodation: 44.8 percent. Food: 24.8 percent. Shopping: 12.1 percent. Entertainment: 8.7 percent.Share of total tourism expenditure for domestic and international visitors by category. Source: Figure created by authors using data from Bank of Italy International Tourism survey an ISTAT Trips and Holidays survey, 2019
The two side-by-side pie charts show domestic and international visitor spending across five expenditure categories. The caption lists the categories as: “Local transportation”, “Accommodation”, “Food”, “Shopping”, and “Entertainment”. Data for each pie chart is provided in a clockwise sense. Domestic visitors (clockwise): Local transportation: 36.3 percent. Accommodation: 29.6 percent. Food: 20.7 percent. Shopping: 12.4 percent. Entertainment: 1.1 percent. International visitors (clockwise): Local transportation: 9.6 percent. Accommodation: 44.8 percent. Food: 24.8 percent. Shopping: 12.1 percent. Entertainment: 8.7 percent.Share of total tourism expenditure for domestic and international visitors by category. Source: Figure created by authors using data from Bank of Italy International Tourism survey an ISTAT Trips and Holidays survey, 2019
Estimated coefficients of the regression model for average expenditure by category (EUR) per person per day of international visitors
| Local transportation | Accommodation | Food | Shopping | Entertainment | |
|---|---|---|---|---|---|
| January | −1.290 | −21.220*** | −5.870** | 3.020 | 1.100 |
| February | −1.190 | −23.160*** | −3.490 | 2.650 | 2.330 |
| March | −3.290*** | −23.000*** | −2.170 | 3.270 | 4.480*** |
| April | 1.110 | −10.280** | 2.000 | 1.520 | 4.330*** |
| May | 0.750 | −10.880*** | 4.390* | 7.820*** | 9.320*** |
| June | 1.100 | −0.520 | 4.260* | 3.520 | 1.210 |
| July | 1.500 | 2.700 | 0.200 | −1.510 | −0.660 |
| September | 1.470 | −22.330*** | 2.800 | 3.480 | 0.380 |
| October | 3.870*** | −26.660*** | 9.510*** | 2.630 | 2.010 |
| November | 3.490*** | −26.470*** | 3.020 | 6.160* | 2.340 |
| December | −1.210 | −11.720** | −1.350 | 5.770 | −6.120*** |
| Manager | 3.470*** | 9.670*** | 3.940** | 2.650 | 3.430*** |
| Industrial worker | −6.930*** | −36.590*** | −17.800*** | −9.180* | −12.690*** |
| Self-employed | −1.250 | −15.070*** | −2.440 | 8.590*** | −7.220*** |
| Student | −1.960 | −26.840*** | −11.750*** | −4.380 | −7.070*** |
| Retired | −7.680*** | −44.110*** | −11.490** | −23.090*** | −15.230*** |
| Other | −2.110*** | −11.330*** | 4.280*** | −4.650** | −5.740*** |
| Age 15–24 | −14.710*** | −74.520*** | −24.190*** | −22.740*** | −22.550*** |
| Age 25–34 | −11.140*** | −64.920*** | −20.440*** | −17.750*** | −20.220*** |
| Age 35–44 | −5.530*** | −30.000*** | −9.730*** | −8.570*** | −10.160*** |
| Age 65+ | −1.500 | 6.970 | −14.970*** | 2.480 | 1.320 |
| Constant | 32.090*** | 175.880*** | 76.240*** | 41.680*** | 40.190*** |
| Obs. | 9,176 | 9,176 | 9,176 | 9,176 | 9,176 |
| Local transportation | Accommodation | Food | Shopping | Entertainment | |
|---|---|---|---|---|---|
| January | −1.290 | −21.220*** | −5.870** | 3.020 | 1.100 |
| February | −1.190 | −23.160*** | −3.490 | 2.650 | 2.330 |
| March | −3.290*** | −23.000*** | −2.170 | 3.270 | 4.480*** |
| April | 1.110 | −10.280** | 2.000 | 1.520 | 4.330*** |
| May | 0.750 | −10.880*** | 4.390* | 7.820*** | 9.320*** |
| June | 1.100 | −0.520 | 4.260* | 3.520 | 1.210 |
| July | 1.500 | 2.700 | 0.200 | −1.510 | −0.660 |
| September | 1.470 | −22.330*** | 2.800 | 3.480 | 0.380 |
| October | 3.870*** | −26.660*** | 9.510*** | 2.630 | 2.010 |
| November | 3.490*** | −26.470*** | 3.020 | 6.160* | 2.340 |
| December | −1.210 | −11.720** | −1.350 | 5.770 | −6.120*** |
| Manager | 3.470*** | 9.670*** | 3.940** | 2.650 | 3.430*** |
| Industrial worker | −6.930*** | −36.590*** | −17.800*** | −9.180* | −12.690*** |
| Self-employed | −1.250 | −15.070*** | −2.440 | 8.590*** | −7.220*** |
| Student | −1.960 | −26.840*** | −11.750*** | −4.380 | −7.070*** |
| Retired | −7.680*** | −44.110*** | −11.490** | −23.090*** | −15.230*** |
| Other | −2.110*** | −11.330*** | 4.280*** | −4.650** | −5.740*** |
| Age 15–24 | −14.710*** | −74.520*** | −24.190*** | −22.740*** | −22.550*** |
| Age 25–34 | −11.140*** | −64.920*** | −20.440*** | −17.750*** | −20.220*** |
| Age 35–44 | −5.530*** | −30.000*** | −9.730*** | −8.570*** | −10.160*** |
| Age 65+ | −1.500 | 6.970 | −14.970*** | 2.480 | 1.320 |
| Constant | 32.090*** | 175.880*** | 76.240*** | 41.680*** | 40.190*** |
| Obs. | 9,176 | 9,176 | 9,176 | 9,176 | 9,176 |
Note(s): Coefficients: * significant at 10%, ** significant at 5%, *** significant at 1%. The constant refers to the average expenditure of a representative tourist, who is an employee between 45 and 64 years old, visiting Rome in August 2019. All regressors are binary variables (dummy) and their coefficients can be interpreted as the difference in average expenditure between the category they represent and the representative tourist
Source(s): Table created by authors
The last element we need for our estimate is the length of stay of both domestic and international visitors. To estimate this, we used data from the two surveys previously described, which revealed that the average length of stay for international holiday visitors to Rome was about 11 nights in 2019 (probably due to the large number of other attractions available in Rome), while the average length of stay for domestic visitors was about 3 nights. Therefore, we decided to take a conservative approach in estimating tourism expenditure by assuming that the length of stay in Rome for international visitors that can be attributed directly to the Colosseum is in line with that of domestic visitors (i.e. 3 nights).
Reasonably, we can also consider 3 nights as the minimum stay required by an international visitor to visit a place or participate in an event in a foreign country. Based on the above assumptions, we estimated the tourism expenditure. Moreover, to avoid double counting with respect to the economic contribution from the direct operation of the Colosseum, per person expenditure for the entrance fee, related services (audio guide, guided tour, etc.), and souvenirs purchased at the bookshop are subtracted from the estimated average expenditure for, respectively, cultural and entertainment services (“Entertainment”) and the purchase of goods in stores (“Shopping”). The estimated size of this overlap in tourism expenditure is about 25.5 EUR/mln. At the end of this section, we also provide a sensitivity analysis related to our main assumptions. Figure 2 shows the estimated total tourism expenditure by category that can be attributed to the Colosseum.
The pie chart shows spending amounts across five expenditure categories. The caption lists the categories as: “Local transportation”, “Accommodation”, “Food”, “Shopping”, and “Entertainment”. Data is provided in the clockwise sense. Local transportation: 172.8. Accommodation: 500.3. Food: 286.2. Shopping: 145.2. Entertainment: 86.4.Total induced tourism expenditure of Colosseum’s visitors by category. Total expenditure is equal to 1,190.9 EUR/mln. Source: Figure created by authors using data from Bank of Italy International Tourism survey an ISTAT Trips and Holidays survey, 2019
The pie chart shows spending amounts across five expenditure categories. The caption lists the categories as: “Local transportation”, “Accommodation”, “Food”, “Shopping”, and “Entertainment”. Data is provided in the clockwise sense. Local transportation: 172.8. Accommodation: 500.3. Food: 286.2. Shopping: 145.2. Entertainment: 86.4.Total induced tourism expenditure of Colosseum’s visitors by category. Total expenditure is equal to 1,190.9 EUR/mln. Source: Figure created by authors using data from Bank of Italy International Tourism survey an ISTAT Trips and Holidays survey, 2019
Following our approach, by combining the average tourism expenditure and the number of visitors of the Colosseum that would not otherwise visit Rome, we derive an estimate of the level of tourism expenditure directly attributable to the Colosseum.
3.3 Indirect use value
Here, the indirect use of the Colosseum is not captured by an economic or financial transaction. However, people derive welfare from living in a historical environment, and this welfare is reflected in real estate prices observed in the housing market. In this section, we provide an estimate of the indirect use value of the Colosseum related to housing comfort.
To achieve this, we employ the hedonic pricing method, an indirect approach that seeks to elicit preferences from observed market-based information. By applying this method, it is possible to estimate the marginal implicit price of the attribute of interest, which provides the additional amount of money that must be paid by an individual to purchase an identical market good with a higher level of that attribute.
To estimate the indirect use value of the Colosseum, we defined a regression model for the price per square meter as a function of elements relevant to potential buyers’ decisions regarding dwellings located in the area surrounding the Colosseum. Data related to housing market transactions have been collected for different radial distances. The collected data includes the announcement date, type of contract, distance from the Colosseum, year of construction, square meters, number of rooms, car parking, floor, total number of floors in the building, presence of a lift, type of property, view of the Colosseum, price, and heating expenses.
As previously mentioned, a functional relationship is specified between the market price and all the relevant attributes of the market commodity. The most common form is the semi-logarithmic form, which has the advantage that the coefficient estimates are proportions of the price that are directly attributable to the respective characteristics. The advantage of the log-log form is that the estimates are elasticities relative to each considered characteristic. In general, this relationship may be described with the use of the following function:
where:
Xi is the vector of the characteristics of the good i;
αi is the vector of parameters;
εi is the disturbance term (Belniak and Wieczorek, 2017).
The variable of main interest is an indicator for the apartment having a view of the Colosseum. Not surprisingly, indicators for the Colosseum view, car parking availability, excellent condition, and top floor have a statistically significant positive effect on house price. On the other hand, all else equal, houses located more than 800 metres away from the Colosseum, and in lower floors have a statistically significant lower price. Specifically, after controlling for other characteristics, we estimate that the house prices with Colosseum view are about 17.5% higher (Table 4).
Estimated coefficients of the model for the logarithm of price
| Colosseum view | 0.175 | *** |
| Distance>800 mt | −0.146 | *** |
| Car parking | 0.187 | ** |
| Basement | −0.755 | *** |
| Ground floor | −0.287 | *** |
| Floor 1 | −0.179 | *** |
| Floor 2 | −0.101 | |
| Floor 4+ | −0.098 | |
| Lift | −0.043 | |
| To be refurbished | −0.111 | * |
| Just refurbished | 0.105 | ** |
| Top floor | 0.155 | ** |
| Constant | 8.844 | *** |
| Obs. | 256 | |
| R-sq | 0.332 |
| Colosseum view | 0.175 | *** |
| Distance>800 mt | −0.146 | *** |
| Car parking | 0.187 | ** |
| Basement | −0.755 | *** |
| Ground floor | −0.287 | *** |
| Floor 1 | −0.179 | *** |
| Floor 2 | −0.101 | |
| Floor 4+ | −0.098 | |
| Lift | −0.043 | |
| To be refurbished | −0.111 | * |
| Just refurbished | 0.105 | ** |
| Top floor | 0.155 | ** |
| Constant | 8.844 | *** |
| Obs. | 256 | |
| R-sq | 0.332 |
Note(s): Coefficients: * significant at 10%, ** significant at 5%, *** significant at 1%. The constant refers to the average log price per sqm of a representative house, that is a 3rd floor house within 800 mt from the Colosseum, without car parking and lift and with no Colosseum view. All regressors are binary variables (dummy) and their coefficients can be interpreted as an approximation of the percentage difference in average price between the category they represent and the representative house
Source(s): Table created by authors
Then, we estimate the impact of the Colosseum view as the average partial effect of the Colosseum view. The result provides a measure of the additional price per square metres due to the presence of the view. This value has been used to estimate the value of the view of the Colosseum taking into account the number of sqm of buildings located in the surrounding area.
From our model, we estimate the implicit price of living in the proximity to the Colosseum as the marginal effect of the view of the Colosseum on the house prices, after controlling for all other relevant characteristics of the houses. The reason why we consider only the extra value of a house specifically enjoying the view of the Colosseum, instead of the higher extra value of simply living in its proximity, is to avoid overestimating the implicit price of living near the Colosseum due to the presence of some other historical attractions in the same area. The result provides a measure of the additional value per square meters (sqm) due to the view.
3.4 Existence value
People may value the Colosseum as “iconic” or “symbolic” or may value the contribution of the Colosseum to its national culture. Non-use values, or passive use values, must be considered in contexts where an individual is willing to pay for a good, even though they do not make a direct use of it, does not directly benefit from it and cannot plan any future use for themselves or others. Thus, since the economic contribution outlined in the previous section may not fully reflect the total contribution of the Colosseum to the welfare of the Italian community, in this section we also estimate the existence value of the Colosseum.
The most common method to quantify the non-use value of heritage is by estimating individual willingness to pay for its preservation through a stated preference technique such as contingent valuation (CV) and choice modelling (CM). These techniques are based on a survey conducted among a representative sample of the target population potentially interested in the heritage element. Each individual in this sample is asked to reveal preferences about his or her maximum willingness to pay (WTP) to secure a public service or avoid its loss or deterioration.
Applied to a heritage asset like the Colosseum, stated preference techniques enable to estimate the economic value the society gives to the existence of the heritage. To this aim, we developed and fielded a survey of Italian residents and international visitors to Rome. This survey was administered online to a sample of 850 Italian residents and designed to reveal attitudes towards the Colosseum. In order to correct for survey design and non-response problems, survey results are weighted to reflect the overall Italian population. To estimate the non-use value, economists apply methods of measuring individual preferences, such as Contingent Valuation Method (CV) and Choice Modelling (CM). In particular, CV seeks measures of the WTP through direct questions (such as “What are you willing to pay?” and “Are you willing to pay X USD?”), while CM seeks to secure rankings and ratings of alternatives from which WTP can be inferred. CM may avoid some of the response difficulties that can be found in CV: for example, dichotomous choices in CV may be subject to “yea-saying”, with respondents giving affirmative but probably false responses; CM avoids this limitation because it provides respondents with a plurality of chances in the interview to express a preference for the asset to be valued over a range of payment amounts (Pearce and Özdemiroǧlu, 2002).
To estimate the existence of the Colosseum we adopt the CV method [1], by asking respondents how much they would allocate to fund the preservation of the Colosseum. Since open-ended elicitation is known to have several limitations and drawbacks (large non-response rates, protest answers, zero answers and unrealistically large bids, unreliable responses) we used a payment card elicitation that provides a context to the bids, while avoiding starting point bias and reducing the number of outliers. Specifically, we asked the following question: “If no action is taken, the Colosseum is expected to deteriorate over time. Which of the amounts listed below best describes your maximum willingness to pay every year to fund the preservation of the Colosseum?”. Then, we provided respondents with a visual aid containing a number of monetary amounts [2]. Unfortunately, beside these advantages, this type of elicitation leads to a censoring of the willingness to pay of the respondents. We used an interval regression model to estimate the average WTP which is a generalization of the Tobit model because it extends censoring beyond left-censored data or right-censored data to the case of interval-censored data. The values of the outcome variable may be either observed (point data) or unobserved but known to fall within an interval (interval-censored data) with the following model equation:
where yj is a continuous outcome, xj the regressors and β the corresponding coefficients. The model assumes that the error term is normally distributed, i.e. εj N (0, σ2). For observations j C, we observe yj, that is, point data. Observations j I are intervals, and we only know that the unobserved yj is in the interval [y1j, y2j]. The likelihood for these censored observations contains terms of the form Pr(Yj ≤ yLj) for left-censored data and Pr(Yj ≥ yRj) for right-censored data. Thus, the parameters of the model β can be estimated using the maximum likelihood method.
4. Results
4.1 Economic contribution: direct, indirect and induced effect
The direct cash flows produced from the operation of the Colosseum corresponds to about 75.3 EUR/mln of direct tourism expenditure by its visitors in 2019. These revenues are mainly given by the expenditure for entrance fees paid from the tourists entering the Colosseum and for additional services related to the visit such as guided tours, purchase of audio guides, ticket pre-sale rights, and royalties on gadgets sold at the bookshops. Overall, the operation of the Colosseum generates about 63.3 EUR/mln of value added. Our estimate of the indirect and induced contribution is 37.5 EUR/mln of value added.
The total yearly contribution is then given by the sum of direct, indirect and induced effects. We estimate that the operation of the Colosseum generates 100.8 EUR/mln in total value added to the Italian economy.
4.2 Induced tourism
Our estimates, indicate that the Colosseum was responsible for about 1,190.9 EUR/mln of tourism expenditure in 2019 (Table 5). As for the expenditure related to direct visit to the Colosseum, this expenditure must also be converted into value added. Tourism expenditure attributed to the Colosseum is estimated to contribute to the Italian economy about 696.9 EUR/mln in direct value added and about 592.7 EUR/mln in indirect and induced value added. Thus, our estimate of the total value added is about 1,289.6 EUR/mln.
Estimated economic contribution of operation of the Colosseum
| Direct | Indirect and induced | Total | |
|---|---|---|---|
| Expenditure (EUR/mln) | 75.3 | ||
| Value added (EUR/mln) | 63.3 | 37.5 | 100.8 |
| Employment (FTE) | 155 | 1,062 | 1,217 |
| Direct | Indirect and induced | Total | |
|---|---|---|---|
| Expenditure (EUR/mln) | 75.3 | ||
| Value added (EUR/mln) | 63.3 | 37.5 | 100.8 |
| Employment (FTE) | 155 | 1,062 | 1,217 |
Source(s): Table created by authors
For the sensitivity analysis were reported our main assumptions on the economic contribution of tourism induced by the Colosseum, namely the percentage of visitors of the Colosseum that would not otherwise visit Rome and their average length of stay (Table 6). The maintained assumptions are in bold. The three numerical rows show the impact of different average length of stay, from only one night to the actual number of nights observed in the sample. The three numerical columns of the Table show the impact of the percentage of visitors of the Colosseum that would not otherwise visit Rome, or at least would not stay for the same length of time. This percentage ranges from 20% to 40%, where 28% is the percentage estimated from the survey. Varying our assumptions would mean that the economic contribution of induced tourism is between 271 EUR/mln and 1,791 EUR/mln.
Estimated economic contribution of tourism induced by the Colosseum
| Direct | Indirect and induced | Total | |
|---|---|---|---|
| Expenditure (EUR/mln) | 1190.9 | ||
| Value added (EUR/mln) | 696.9 | 592.7 | 1289.6 |
| Employment (FTE) | 24,693 | 16,791 | 41,483 |
| Direct | Indirect and induced | Total | |
|---|---|---|---|
| Expenditure (EUR/mln) | 1190.9 | ||
| Value added (EUR/mln) | 696.9 | 592.7 | 1289.6 |
| Employment (FTE) | 24,693 | 16,791 | 41,483 |
Source(s): Table created by authors
We report a summary of the overall contribution of the Colosseum (Table 7).
Sensitivity of economic contribution of tourism induced by the Colosseum (expenditure EUR/mln) to varying length of stay and percentage of visitors of the Colosseum that would not otherwise visit Rome
| Average length of stay | Percentage of visitors attributed to the colosseum | ||
|---|---|---|---|
| 20% | 28% | 40% | |
| 1 | 271.4 | 380.0 | 542.8 |
| 3 | 850.6 | 1190.9 | 1701.2 |
| Observed | 895.4 | 1253.6 | 1790.9 |
| Average length of stay | Percentage of visitors attributed to the colosseum | ||
|---|---|---|---|
| 20% | 28% | 40% | |
| 1 | 271.4 | 380.0 | 542.8 |
| 3 | 850.6 | 1190.9 | 1701.2 |
| Observed | 895.4 | 1253.6 | 1790.9 |
Note(s): Central case in italic
Source(s): Table created by authors
4.3 The implicit price living in the proximity of the colosseum
The estimated value of the Colosseum view is 1,232 EUR per square meter. Then, we multiply this unit value by the number of total square meters of houses located in the surrounding area (called Archeological area). As shown in Table 8, our estimate of the indirect use of the Colosseum related to housing comfort is 406.1 EUR/mln.
Indirect use value of the Colosseum
| Unit extra value that can be attributed to the colosseum (EUR/sqm) | Houses located in the surrounding area (sqm) | Indirect use value (EUR/mln) |
|---|---|---|
| 1232.2 | 329,584 | 406.1 |
| Unit extra value that can be attributed to the colosseum (EUR/sqm) | Houses located in the surrounding area (sqm) | Indirect use value (EUR/mln) |
|---|---|---|
| 1232.2 | 329,584 | 406.1 |
Source(s): Table created by authors
4.4 Existence value
We found that the average willingness to pay is higher for residents in Rome than for other Italians, but the difference is not statistically significant (Table 9).
Estimated coefficients of the interval regression model for stated preferences over the preservation of the Colosseum
| Rome – rest of Italy | 32.667 |
| Rest of Italy | 56.889** |
| lnsigma | 5.134*** |
| Obs. | 850 |
| Log-lik. | −351.4 |
| Rome – rest of Italy | 32.667 |
| Rest of Italy | 56.889** |
| lnsigma | 5.134*** |
| Obs. | 850 |
| Log-lik. | −351.4 |
Note(s): *significant at 10%, **significant at 5%, ***significant at 1%
Source(s): Table created by authors
According to our estimates, on average the respondents suggested a yearly funding of 58.4 EUR per person for the preservation of the Colosseum (Table 10). Note that the estimated WTP is higher for residents in Rome (89.6 EUR) than for people living elsewhere in Italy (56.9 EUR). Nationwide, considering the 18 years or older Italian residents, this is equivalent to about 2,936 EUR/mln per year. As highlighted previously, the estimation of the existence value of the Colosseum should be evaluated in the Italian context, characterized by the cultural heritage sites and artistic events exploitation and large audience participation.
Willingness to pay for the preservation of the Colosseum (contingent valuation method)
| Average WTP | Total WTP | |
|---|---|---|
| (EUR) | (EUR/mln) | |
| Rome | 89.6 | 212.6 |
| Rest of Italy | 56.9 | 2723.2 |
| Italy | 58.4 | 2935.9 |
| Average WTP | Total WTP | |
|---|---|---|
| (EUR) | (EUR/mln) | |
| Rome | 89.6 | 212.6 |
| Rest of Italy | 56.9 | 2723.2 |
| Italy | 58.4 | 2935.9 |
Source(s): Table created by authors
5. Conclusions
In this paper, we emphasize how the preservation of cultural assets contribute to improving local economic conditions, generating a positive impact on the local economy and on the amount of welfare that such heritage assets generate for the society. Specifically, we provided an analysis and quantification of the economic and social value of the Colosseum, with a focus on its impact on the Italian economy and society. The economic value of a cultural heritage site like the Colosseum can be defined as the amount of welfare that such a heritage generates for society. Thus, we looked at different aspects, such as its main economic contribution related to its direct use, its indirect use value, and its non-use value related to non-market aspects, capturing its existence value.
Considering both its direct operation and induced tourism, we estimated that the Colosseum generates about 1,390.5 EUR/mln a year in terms of value added to the Italian economy (Table 11). In addition to the economic contribution from its direct use, using a hedonic pricing model we estimated that the indirect use value of the Colosseum, in terms of housing comfort for residents, is about 406 EUR/mln.
Estimated total economic and social value of the Colosseum
| Components | Annual values | Asset values |
|---|---|---|
| (EUR/mln) | (EUR/mln) | |
| Economic contribution | 1,390 | |
| Direct Economic Contribution | 63.3 | |
| Indirect Economic Contribution | 37.5 | |
| Tourism | 1289.6 | |
| Indirect use value | 406 | |
| Existence value (non-use value) | 2,936 | 75,672 |
| Components | Annual values | Asset values |
|---|---|---|
| (EUR/mln) | (EUR/mln) | |
| Economic contribution | 1,390 | |
| Direct Economic Contribution | 63.3 | |
| Indirect Economic Contribution | 37.5 | |
| Tourism | 1289.6 | |
| Indirect use value | 406 | |
| Existence value (non-use value) | 2,936 | 75,672 |
Source(s): Table created by authors
Note that this is not an annual figure. Finally, using a contingent valuation method, our estimates suggested that the Colosseum’s non-use value, capturing its existence value, is about 2,936 EUR/mln per year. Different from the indirect use value, the existence value is an annual figure. Thus, to convert these figures to an asset figure, we took a 100-year present value with a discount rate of 4.0% a year.
Thus, we estimated the total existence value of the Colosseum at a 100-year present value of 75,672 EUR/mln.
6. Implications
This research holds significant implications for both scholars and practitioners in the fields of cultural heritage valuation and management. For scholars, it offers an innovative methodology for conducting comprehensive economic evaluations of cultural assets, paving the way for more effective and sustainable management practices. The capacity to quantify both tangible and intangible well-being values associated with these sites establishes a strong foundation for shaping preservation and enhancement policies. Moreover, this study lays a solid foundation for future research endeavors, motivating academics to further explore the field of assessing economic and social value within the realm of cultural heritage.
The results of our studies can be integrated into a cultural heritage management framework composed of a comprehensive, multidisciplinary framework for confronting the challenges of cultural heritage management. Roxas (2020) developed a multidisciplinary framework to address the challenges of cultural heritage management, emphasizing the importance of stakeholder engagement, sustainability, and regulation. Our results provide valuable insights that align with this framework, especially fostering an economic valuation in guiding effective decision-making. By incorporating our findings, cultural heritage managers and policymakers can enhance their ability to make well-informed choices that contribute to the preservation and sustainable management of cultural assets encompassing diverse dimensions and the interests of all stakeholders involved.
In particular, given the cultural and artistic accessibility-oriented policy run by the Italian policy maker, the results of the present paper and the several cultural asset heritage dimensions highlighted should be considered of greater value to assist and evaluate the most appropriate cultural asset management strategy.
The views and opinions expressed in this study are solely those of the authors. They should not be interpreted in any way as reflecting the perspectives of the institutions which the authors represent.
The views and opinions expressed in this study are solely those of the authors. They should not be interpreted in any way as reflecting the views of the Institutions to which the authors belong.
Notes
An established result in the literature on methods to elicit preferences, is that the willingness to pay is typically lower (even much lower) when estimated using the CV relative to estimates from the CM. Such a divergence of WTP estimates is actually due to the fact that the CV estimates will tend to be biased downward, and the CM estimates will tend to be biased upward (see, for example, Carson et al., 1999; Boyle et al., 1996). As an alternative method to estimate the WTP, we also estimated the WTP for the preservation of the Colosseum using the CM approach, building upon the specification of an indirect random utility. Results are available from authors upon request.
The full stated preference survey questionnaire is available from authors upon request.
