«Holger Kächelein Endrit Lami Drini Imami Working Paper No. 71 April 2010 b B A M B AMBERG E CONOMIC R ESEARCH GROUP k k* BERG Working Paper Series ...»
Elections Related Cycles in Publicly Supplied Goods
Working Paper No. 71
BERG Working Paper Series
on Government and Growth
Bamberg Economic Research Group on Government and Growth Bamberg University Feldkirchenstraße 21 D-96045 Bamberg Telefax: (0951) 863 5547 Telephone: (0951) 863 2547 E-mail: firstname.lastname@example.org http://www.uni-bamberg.de/vwl-fiwi/forschung/berg/ ISBN 978-3-931052-79-9 Reihenherausgeber: BERG Heinz-Dieter Wenzel Redaktion Felix Stübben∗ ∗ email@example.com Elections Related Cycles in Publicly Supplied Goods in Albania Holger Kächelein Faculty of Business Administration and Economics, University of Tirana DAAD-Lecturer Economics firstname.lastname@example.org Endrit Lami Part time Lecturer, Faculty of Business Administration and Economics, University of Tirana email@example.com Drini Imami Faculty of Economics and Agribusiness, Agriculture University of Tirana Institute of Advanced Studies, University of Bologna Open Society Institute - GSGP Grantee firstname.lastname@example.org Abstract The phenomena of m anipulation of the economy by th e incumbent for electoral purpose are called Political Business Cycles (PBC), introduced by Nordhaus (1975). Using policy control economic instruments, as fiscal and m onetary instruments, government may manipulate the economy to gain electoral advantage by producing growth and decreasing unemployment before elections.
In addition to increased public expe nditures, also the produc tion/supply of certain publicly provided goods m ay score im provements. In Alba nia, production and supply of electricity (for the time span of our a nalyzes) was contr olled by K ESH (Korporata Energjitike Shqiptare – Al banian Energy Corporation) whic h is a quasi- m onopoly in the supply of electricity in Albania, and it is publicly run. Thr oughout the transition, supply of electricity, due to various technical and economic reasons, has not been stable, and characterized by syst ematic interruption for house holds and businesses users, affecting their well-being and perform ance (electricity is a m ain source of energy for households, including heating a nd cooking). Therefore, it seems so that there is an incentive and rationale for the incumbent to use also the provision of elect ricity to impress the voters before elections, beside of the classical instruments of expenditures.
In this paper we analyze consum ption, production and import of electricity in Albania.
Our hypothesis is that before elections, elec tricity consumption may increase above usual levels, followed by a contractio n after elections. In our analys is we use m odern standard econometric approach, used widely for research related to PBC. By ARMA modelling it is possible to prove if elections can explain cha nges in electricity prod uction, in addition to the past history of the variable and the random error term.
Keywords: Political Business Cycle, Electricity, Albania JEL classification code: P26, E32, D72, H72
HOLGER KÄCHELEIN, ENDRIT LAMI AND DRINI IMAMIElections Related Cycles in Publicly Supplied Goods in Albania
1. Introduction It seems to be obvious t hat the economic performance of a governmen t determines to a large extent its likelihood of reelection as confirmed by Fair (1978, 1982, 1988), Madsen (1980) or Lewis-Beck (1988), and therefore econom ic factors influence political factors and the other way around. Furthermore, incumbents may use their power and the instruments available to influence the economic environment especially prior to election to im prove the likelihood of re election. Over the last decades, there has been plenty of res earch and articles pub lished on such an opportunistic behaviour of po liticians, aiming to analy ze and explain the use of fiscal and m onetary instruments by the incum bent to stim ulate economic performance before elections, to im press the voters. The traditional Political Business Cycle (PBC) literature, as introduced by Nordhaus (1975), concentrated on an exploitable Phillips curve, to explain th e use of econo mic instruments to affect macroeconomic variables, such as unem ployment, GDP, etc. Evidence of PBC was also found in several less developed and dem ocratic countries. Gimpelsen (2001) made a research on the existence of PBC in Russia, finding evidence in support of it.
Another study of Asutay (2004) provi ded clear evidence for the presence of PBC in Turkey. The incumbent in Turkey has used fiscal and m onetary policy instruments to create PBC in order to im prove the chances of being reelected. Also previous research on the existence of PBC i n Albania indicated that the incu mbent manipulates fiscal instruments, increas ing public expenditure s before elections, including public i nvestments, expenditure on com pensation of em ployees in parliamentary, social assi stance, while regarding t he macroeconomic outputs, we have found, partial evidence of PBC in GDP and unemployment but not in inflation (Imami and Lami, 2006).
After Nordhaus (1975) ini tial contribution there was an increasing resear ch interest focusing on budget cycles, based on the observations of Tufte (1978) and Frey and Schneider (1978a, b). Even though there is a wide consensus about t he importance of the actual econom ic conditions in pleasing the voters, there is still doubt about the ability to influence the macroeconomic indicator in a precise 4 Holger Kächelein, Endrit Lami, Drini Imami predictable manner. Taking the limitations into account, ne wer approaches focused on pre-election manipulations of fiscal policy instruments. As shown by Brender and Dazen (2005) and Shi and Svensson (200 6), especially new dem ocracies are vulnerable for such political budget cycles. While Alt and Lasse n (2006) show the relevance of transparency, Brender and Dazen (2005) also pronounce the lack of experience that voters have in new dem ocracies regarding the ex istence of political fiscal cycles. Meanwhile, Shi and Sven sson (2006) see beside the aspect of information also the incumbents’ rents of staying in power as a relevant aspect.
However, incumbents may not use only classical instruments as the composition and the size of the public budget if there are also other instruments available. These approaches mentioned above may explain why political budget cycles arise even though those voters should punish suc h behaviour. Another pr oblem related to political budget cycles is the tim ing of the activity. Since the incumbent cannot precisely estimate the lag between the stim ulus as a change in the public budget and the impact on the econom ic environment, they may be interested to use ot her instruments having a m ore direct impact on the economy and the well being of the voters.
We try to shed light on the question, whether incum bents may use other instruments available, beside classical fiscal instruments, to impress voter in years of election. Based on the results that political budget cycles seems to be a phenomenon of developing countries or new dem ocracies, we focus on Albania, a country with a relatively short experience of democracy, which provides only a minimum of fiscal transparency (IBP 2009a, b). In t his paper, we focus specifically on electricity, which is a publicly provi ded good in Al bania and which is characterized by special features. Given that electricity represents one of t he most basic needs, househol ds should be highly sensible concerning sufficient supply of electricity. Furtherm ore, it is quite expensive to storage electricity and only for selected purposes, such as heating or cooking, substi tutes are availabl e and partly used. Furt hermore, in the case of Albania we have a lim ited supply meanwhile de mand has increased dramatically after the system change. And finally, the Al banian electricity market was a quasi public monopoly.
Given that electricity supply (consum ption) relies l argely on both im ports and domestic production, it is important, in this context, to analyze both these sources of electricity – in addition to consumption per se. In our paper we analyze consumption as well as production and im port dynamics of electricity by KESH which is a quasiElections related cycles in the publicly supplied goods in Albania 5 ly run.1 Our monopoly in the suppl y of electricity in Albania, and it is public hypothesis is that before elections, electri city consumption, production and im port may increase above usual le vels, followed by a contraction after elections. In t his paper we focus on the parliamentary elections in 2001 and 2005 – during that period was common to observe electricity supply shortages throughout Albania. In our analysis we use m odern standard economet ric approach, used wi dely for research related to PBC, aim ing to test if elections can explain ch anges in electricity supply in form of production and import.
In the upcom ing chapter we will present a short overview about the electricity provision in Albania, to provi de background information concerning t he existing undersupply as a precondition for using elec tricity supply as an instrument around elections to impress the voters. Chapte r three provides an overview about the method and data used while chapter four presents the main findings.
As percentage of 7.3% 13.1% 13.7% 10.4% 8.5% 9.8% Demand Source: World Bank (2006): p.235, own calculations
In the time span of our analysis, OSSH (Operatori i Sistemit te Shperndarjes – Distribution System Operator) was part of KESH.
6 Holger Kächelein, Endrit Lami, Drini Imami Major problems ar e also the low tariffs which do not cover the costs, network losses and unpaid bi lls. As a results, the Al banian government has had to subsi dize the state own electri city company KESH. In 2005, KESH produced a quasi public deficit of 1.8 percent of the GDP, implying losses to be covered by the public budget (World Bank: 2006: p 25).
There are different reasons for interrup tion of electricity. One of the main reasons is that more than 95% of electri city production, is based on hydro pow er (Nashi 2009), so oscillation in hydro deposit levels, affected by natural factors (rain, draught) directly affect the availability of electricity. The gap, between the demand and production, is partia lly covered by im ports, while the remaining gap, not covered by dom estic production or im ports (for natural, financ ial or technical reasons) is translated into systematic, but oscillating, interruption of electricity.
Turning to household consumption, Alba nians have still suffered under unmet basic needs. In 2002, ba sed on t he non incom e poverty indicators, every third Albanian has to be considered as poor a nd every 10th Albani an as extremely poor.
Indicators as inadequate wate r and sanitation, inade quate housing, crowding or lack of education can only be infl uenced in the longer run. Meanwhile, the supply of electricity can be influenced even in the short run, as th e electricity grid has a bro ad reach and therefore, electricit y could be virtually everywhere available. In 2002, more than 13 percent of the Albanian house holds suffered under power shut offs for 6 hours or more per day (World Bank, 2003: p. 17).
Table 2. Frequency of power supply interruption
Table 2 gives an overview of t he frequency of the i nterruptions based again on the Living Standard Measurement Surv ey (LSMS) of 2002. The tim e without electricity supply varied between m ore than 9 hours in rural areas and 5.6 hours i n Elections related cycles in the publicly supplied goods in Albania 7 the capital Tirana. The situation has im proved in t he following years; however, in 2005 nearly 40 percent still reported daily interruptions of power supply (World Bank, 2007: p. 11).
These irregularities ham per the econo mic development of A lbania as well. In 2002, more than three out of four firm s stated power supply as a problem for their business, which is m ore than three times higher com pared to the South East ern Region. As a result of the insufficient electri city sector, a loss of 2.7 to 5.4 percent of GDP is estimated for 2001-2002. Concerni ng the total costs, we have also to add cumulative investments in ba ckup power supplies, roughly of t he same extend the direct impact, however, spread over several years (World Bank 2006, p. 239-240).
3. Method and Data
3.1 Specifications of Variables, Data and empirical tests Since electricity is a n essential good fo r households and busi nesses, we assume that the incum bent may try to im prove its supply before elections, by increasing production and/or increasing imports. Electricity is an important source of energy in Albania. In addition to the wide use in t he industry, electricity is a m ain source of heating and cooki ng for households. As already discussed before, suppl y of electricity in Albania, is char acterized by systematic interruptions whose effects have been deemed as very negative for development of businesses, especially in some sectors, in addition to direct implications for households’ well-being.
In this research work, we intend to te st for possible statistically significant increase of electricity consumption, produc tion and import before elections, in li ne with the incumbent interest to “please” v oters, in order to incr ease likelihood to be re-elected. The tim e series of production, imports and consum ption of electricity time are on m onthly basis, spanning fr om M1-2000 to M12-2008 (from January 2000 to December 2008), adding up to 96 observations. The unit on w hich the data analysis is based is MW/H. There are two parliamentary elections taking place in this period, namely June 24, 2001 and July 3, 2005.
Following the standard approach in this field, 2 we will app ly the Inte rvention Analysis based on Box and Tiao (1975), a methodology for constructing a statistical model in our study. In this paper we test the hypothesis of the existence of changes in the supply – as production and im ports – as well as consum ption of electricity.