Demand Functions for Cinema in Different Provinces in Iran

In this article, the demand functions for cinema in different provinces are estimated using the Almost Ideal Demand System. Then price and income elasticity are calculated for each province and for the entire country. The obtained results show that price elasticity for the entire country is equal to -0.91; and in the provinces except for Isfahan, Hormozgan, Kerman, Kurdistan, Mazandaran, and Eastern Azerbaijan, it is less than 1, which implies price inelasticity. Price elasticity crossed between journals and cinema, in each of the 9 provinces of Kohgilooyeh va Boyrahamd, Chahar Mahalo Bakhtiary, Isfahan, Ilam, Fars, Gilan, Hamedan, Yazd, and Zanjan, is negative, that means the goods could substitute for each other in consumption. But, in other provinces, the elasticity is positive; thus, the goods complement each other in consumption. The elasticity for the entire country is equal to -0.12. The price elasticity crossed between books and cinema for the entire country is -47%. Also, it is negative for each province, which means the two commodities substitute for each other in consumption. The income elasticity for all the provinces is positive and stands between zero and one; and for the entire country it is 57%, which means going to movies is considered a necessary commodity.


Introduction
The examination of household consumption from a demand point of view helps to respond to questions such as: "What factors, and up to what degree, affect the demand for a commodity?"and "What is the degree of sensitivity of demand for a commodity in proportion to changes in household income, in commodity price, or in the price of other commodities?"Finding responses to these questions not only clarifies the consumption patterns of the past, but also helps to predict future trends.Examination of demand for and consumption of cultural commodities including cinema, in Iran, is significant from different angles.First, in recent years the rate of economic growth in Iran has been increasing; a year ago it reached 6.5%, which in future would cause a demand increase for all goods, including cultural commodities.Second, in Iran no movies are produced unless either the government has produced them; has supported their production; or has issued a permit for their production.Therefore, the study of demand and discovering the degree of influence of different factors on demand helps the policymakers to adopt the required policies to respond to demand growth in future, and since there has been no previous study on household demand for movie going in Iran, the necessity of the study is evident.

Objective, Subject and Methodology of Collecting Data and Performing Analysis
The objective of this article is to identify the economic factors that make an impact on demand to go to cinema, and respond to the key question: "What is the demand sensitivity to household income; the ticket price for movies; and the price of other related commodities in different provinces?"To estimate the demand function, the data related to household expenditures in different provinces in , that are collected annually by Iran's Center for Statistics, are used.The provinces under study are: 1-Booshehr; 2-Kohgilooyeh va Boyrahmad; Consumption demand of household is obtained using econometrics techniques and demand equation systems.The applied model is the Almost Ideal Demand System, that is explained in the theoretical framework section.

A survey of movie literature in Iran and other countries
About cinema and film there is an expanse of literature.Advertising films in different countries (Andrew, 1984 ); literature, cinema and their mutual impacts (Cohen,1979;Phillips, 2005); social aspects of movies (Aitken, 1990); the process of the production of movies (Jowett, 1976& Bernstein,1994); advertising films in Russia and Nazi Germany in WWI (Taylor. 1979); the relations between cinema and other mass media such as television; the culture of consumerism in movies; the role of politics in cinema (Doherty, 2005); movie industries in different countries (Armes,1987;Fox, 2000); silent movies (Burows, 2003); identity and policy making in cinema, with consideration of national characteristics (Lee, 2000;Everett, 2005); religious approaches to cinema; cinema circumstances and general perceptions of the 20th century cinema in different countries, including Germany, Japan, and Australia (Yoshimoto, 2000;Rayner, 2000); identity and nationality in British cinema (Cook,1996); paradigms, sociohistorical aspects, conceptualization and policymaking in Indian movies (Chakravarty,1993;Ravis, 2000;Wilson, 2004); a historical survey of cinema in Spain and Portugal, and the role of sex (Faulkner,2004& Mira,2005;Jordan, 2004); popular cinema in Brazil (Shaw, 2004); American policies on cinema at the beginning of the 20th century (Aronstein, 2006;Grieveson, 2005); a historical examination of Italian cinema, the role of cultural issues, social aspects and sex (Bertellini, 2004); French cinema in the 1970's (Hammond, 2000); the present situation of German cinema (Flinn, 2004); a survey of music and style in the movies of West Germany (Berghahn, 2005); comedy and drama in Latin American cinema (Mather, 2005); an examination of movie copyrights in European Union (Kamina, 2002); the role of women in movies (British Film Institute, 1988); Cold War, Post Modernism, Realism, and policymaking in Hollywood (King, 2002;Foertsch, 2001;Scott, 2000); popular cinema in India (Raminder, 2005) are among the issues that are reflected extensively in the literature of cinema.The variety and extensiveness of the subject is not limited to the above.The discussion of national cinema in different countries such as China, Taiwan and Hong Kong in recent years (Zhang, 2004), issues related to urban studies, feminism and the magic ambiance of cinema (Loughline, 2004;Moor, 2005;McCabe, 2004;Chanan, 2004); modernity and cinema (Stewart, 2005); film as a social reality (Turner, 1999); violence and identity in cinema (Becerra, G. 2004); religion and native culture in movies (Griffiths, 2002); new changes in cinema (McCrisken, 2005); entity of cinema (Bazin, 2005); small legends and cultural issues in cinema (Martin, 2001) are other issues to which more attention has been paid in recent years.In fact, so far there is no available comprehensive research about Iranian cinema and particularly its economic aspects, and household demands for movie going.

Theoretical Framework: Almost Ideal Demand System
In the 1980's, for the first time, Deaton and Muellbauer introduced the "Almost Ideal Demand System" (Deaton & Muellbauer, 1980).This model, just like some other models of demand systems, starts with some expenditure functions, known as PIGLOG 1 expenditure functions.The general form of this expenditure function is:

4-1
In which: C= Total expenditure; U= Utility index; P= Price vector According to the theory, consumption is a function of utility level and the price vector that is shown as: C= C(U, p) Also, a(P) and b(p) are functions of price levels.The main characteristic of the expenditure function is its inclusion of expenditure that not only presents the obtainable expenses to reach two levels of the least livelihood and the most welfare, but also includes all the points in between the two levels.Since, according to the theory of consumer behavior, the function of consumer expenditures in proportion to the level of prices shows a homogeneity of the degree one, therefore a(p) and b(p), that are both function of the prices, must be selected so that the results of the algorithm C(U,p), that is a linear composition of the algorithms a(p) and b(p), turn to a homogeneity of degree one.Hence, Deaton and Muellbauer declared Lna(p) and Lnb(p) as: If we multiply both sides at C p i then we get:

4-5
In which W i is the budget share of the ith commodity.Now we take the derivative of the equation 4-4 to the ln p i .Thus, we divide the expression to 3 components so that: And therefore: But the important point is that the derived function of the share above and its corresponding demand function are in fact compensated functions because they are derived from Shephard's Lemma.Also, the general form of the function 4-6 expresses the same point.Since it is a function of the level of prices and utility, this compensated function must be transformed to a corresponding function of un-compensated demand.If, from the above relation, we derive U based on M and p, then we reach the indirect function of utility, and then we get the uncompensated function.In order to do that, using the relation 4-4, we calculate U in respect to M and p, and include that in the function 4-6 to reach the intended results.Thus: With inclusion of this expression in function 4-6 we get: If we declare the bracketed expression in ln p, then: The above expression is the Almost Ideal Demand System of equations.The important point in this system is that considering the price index p the coefficients of the above equation are nonlinear, and to estimate the coefficients non-linear methods must be applied.This requires sufficient data and statistics.Therefore, in most empirical studies, instead of using the real index of p and non-linear method, the Stone index is used as a substitute for the real index p, and thus the model is turned linear and could be easily estimated using linear methods.Related case studies confirm that estimations using the Stone index and the real index p have little difference, hence, it is recommended to use the Stone index in empirical studies.The Stone index is declared as: The amount of p 0 is calculated for each period and is inserted in the model as a specified amount.The relation 4-8 is known as Linear Approximate AIDS (LA-AIDS).The most important reason to select the Almost Ideal Demand System of equations is that the estimation of the model as an approximation is very simple, and the existence or non-existence of the homogeneity and symmetry restrictions can be tested, and since our data and statistics are based on household expenditures, it's fitting based on the household expenditure does not alter the functional form of the model.
the index price associated to Kth commodity group, n is the number of the commodities existing in the system, and all the   , .Then, the PIGLOG expenditure function would be: