Processing the Cometr tax data

Home Based Recycling Business

Make Money in the Recycling Business

Get Instant Access

This section outlines the main stages of processing the tax data set. The aim of this work was to obtain a set of data that could be stored on the E3ME databanks and used in the analysis in the COMETR scenarios.

7.3.1 Model classification

Two new classifications were added to the E3ME model so that it could cope with the new tax data. These were the CT (COMETR Tax) classification, and the CR (COMETR Revenue-Recycling) classification.

The CT classification (see Table 7.1) lists all the taxes that are used in the COMETR scenarios. This is not to say that other taxes were ignored, nor that further detail was not used, but it was possible to express all the taxes as an element in this classification. In addition, specific industry information was added at a later stage where it was available. It should be noted that initially there were 26 elements in the CT classification as energy, CO2 and other emission taxes were separated, but as these taxes are both in terms of EUR/toe energy use (or equivalent measure), they could be simply added together and there was no information gained from keeping them separate. In addition, much of the tax revenue data combined energy, CO2> and other emission tax receipts, so this proved to be a more efficient use of data. The original classification was expanded to include petrol and diesel separately rather than having a single motor spirit entry.

The CR classification (see Table 7.2) handles the various revenue recycling mechanisms employed in the ETRs considered in COMETR.

Table 7.1. The CT classification

1. Industry energy and CO2 tax: Coal

2. Industry energy and CO2 tax: Oil (heating)

3. Industry energy and CO2 tax: Gas

4. Industry energy and CO2 tax: Electricity

5. Industry energy and CO2 tax: Petrol

6. Industry energy and CO2 tax: Diesel

7. Household energy and CO2 tax: Coal

8. Household energy and CO2 tax: Oil (heating)

9. Household energy and CO2 tax: Gas

10. Household energy and CO2 tax: Electricity

11. Household energy and CO2 tax: Petrol

12. Household energy and CO2 tax: Diesel

Source: Cambridge Econometrics.

Table 7.2. The CR classification

1. Reduction in income taxes

2. Reduction in employers' social security contributions

3. Reduction in employees' social security contributions

4. Increase in state benefits (inc pensions)

5. Additional government investment

Source: Cambridge Econometrics.

7.3.2 Rates, revenues, and recycling

7.3.2.1 THREE MAIN TYPES OF INPUT DATA WERE REQUIRED E3ME's energy and environment databanks already contain detailed data on fuel demand, by 19 fuel users and 12 fuels (source: IEA). This forms the base for the tax. The tax rates were compiled, based around the classifications outlined above.

E3ME used the tax rates to estimate the increase in the cost of fuels as a result of the ETR. This was fed into the energy submodel, the fuel demand equations, and then into the rest of the model.

Tax revenues were also required. It would be intuitive to say that tax revenues should be equal to tax rates * fuel use, but this misses out an important part of the modelling (and indeed the focus of one scenario), in that the full tax rates are rarely paid at a macro level. This may be due to special exemptions to specific industries or households, or the failure of central governments to collect tax. In the COMETR reference case, the effective tax rates were determined by the tax revenues divided by fuel use. While this makes some attempt to take into account exemptions and non-payments, it turned out that the data for tax revenues were generally much harder to obtain and are published at a more aggregated level, so assumptions had to be made on a case-by-case basis. See the following sections for more details.

The revenue recycling is an important part of the modelling in E3ME, and in some cases has a larger economic impact than the energy taxes. Therefore it is important that the scenario results demonstrate the most accurate profile of revenue recycling possible, and are not biased by any changes in the overall level of taxation. To achieve this target, the revenue recycling data were converted to shares of revenue received and the shares were set to sum to 1, ensuring that tax receipts equal revenue recycling payments.

All three data sets were obtained in the form of annual time series covering the period 1994-2004. Gaps in the time series (e.g. in 2004) were estimated using linear interpolation or extrapolation based on projections of fuel use and Cambridge Econometrics' custom software algorithms.

7.3.2.2 STANDARDIZED UNITS WERE USED TO STORE THE DATA The main unit of energy data in E3ME is thousands of tonnes of oil equivalent (th toe) and the economic variables are stored and calculated in millions of EUR (tax data are held at current prices, then deflators used to obtain 2000-based series). The units chosen for the COMETR classifications are consistent with this, namely being:

• tax revenues are held in millions of EUR;

• revenue recycling is stored as shares and used to make changes to other taxes in millions of EUR or investment (millions of 2000-based EUR).

In cases where the data provided were in national currency, the exchange rates on E3ME's economic time-series databank were used to convert the data to EUR.

7.3.2.3 THE MAIN CONVERSION FACTORS USED IN THE DATA PROCESSING

These were the main conversion factors used for the energy classifications:

• coal produces 25.4GJ/tonne

• oil produces 43.5GJ/tonne

• gas produces 35.6MJ/CUM

• petrol produces 44.8GJ/tonne

• diesel produces 43.3GJ/tonne

7.3.3 Software inputs

The data were received in a single Microsoft Excel spreadsheet.

The data processing was done using the Ox software package (see <http://www.doornik.com/products.html#Ox>). Ox is a flexible matrix-based software package that has commands similar to those in standard C + + in construct. Ox was used to read in raw data from the spreadsheets, process it, and save it to the E3ME databanks. There was also some interaction between Ox and Visual Basic in accessing the spreadsheet data efficiently and Cambridge Econometrics has its own library of custom Ox software to aid with the processing.

E3ME is programmed in Fortran and controlled by the IDIOM software package. E3ME's direct-access databanks are Fortran-based, but can also be accessed by Ox.

7.3.4 Processing the individual countries

7.3.4.1 INTRODUCTION

This section outlines the main steps and assumptions made in order to process the data for the individual countries as accurately as possible. All data were converted into the units given above using the converters described above. The data for tax rates were usually available, so it was not necessary to make many assumptions during the processing. However, this was not generally the case with revenues and revenue recycling methods. Much of this work drew from the analysis carried out by other partners in the COMETR project.

7.3.4.2 DENMARK

Denmark is different from the other countries in that it makes a clear distinction between 'light' and 'heavy' industry and charges different tax rates to each one. This poses a problem for models such as E3ME that expect a single value. The final methodology counted all industry as heavy, as the majority of emissions fall in this category, but without deductions.

The tax revenues for Denmark were split between energy and CO2 taxes. Although the two were eventually aggregated, this was useful in allocating between fuels and sectors. The energy tax was split by fuel and was not applied to industry (except motor fuels), so this part was straightforward (fuels were shared out assuming that exemptions are equal for households and industry). CO2 taxes were slightly more problematic in that only a national total was available and so these had to be split between households and industry, and also between the different fuel types. With no other information available the sharing was done using fuel tax * fuel uses, that is, the exemptions are the same across households and users. It is unfortunate that no additional information was available to allow this assumption to be relaxed.

Taxes that existed before ETR commenced in 1992 were not included in the analysis. Existing taxes were defined as ones that existed in 1991 and were subtracted from total tax rates to calculate just the ETR part. Likewise, a similar share of the revenues was removed and attributed to existing taxes.

7.3.4.3 GERMANY

Although the tax revenue data for Germany were relatively detailed compared to other countries, the data did not make the distinction between industry and households. Therefore it was necessary to make the following assumptions:

• for motor spirit and electricity, households have the same exemptions as industry;

• for any other fuels, households always pay the full rate.

As household consumption of fuels other than middle distillates and electricity is very low, the second of these assumptions is not particularly important. However, particularly in the case of electricity, assuming the same exemptions are available for households and industry may not be realistic.

Revenue recycling in Germany was relatively complex compared to the other countries, in that three main methods were used. Working on the basis that the cuts in social security contributions were shared equally between employers and employees, the shares were calculated by dividing the extra investment by total revenues, and equally splitting the rest between employers' and employees' social security contributions, so that the shares summed to 1. Existing tax rates were subtracted from the totals, allowing an analysis of just the ETR component of the tax.

7.3.4.4 FINLAND

These data included seven fuels that were aggregated to the fuels in the CT classification. Interestingly, no revenues for gas were included; it is possible that these data were part of light fuel oil but with no information to go on, it was decided to add this separately by calculating tax rate * fuel use (i.e. assuming zero exemptions). As gas use is quite minor in Finland, this assumption should have had little overall bearing on results.

The other major limitation with the Finnish data was that there was no distinction between revenues from industry and households. After reviewing the available literature, it was assumed that households had no exemptions and always paid the full charge. Exemptions for industry were then calculated as total revenue—(industry tax * industry fuel use). Existing taxes in Finland (those that existed in 1996) were subtracted from the totals and were not considered to be part of the 1990s ETR.

There was no specific policy covering revenue recycling in Finland, so it was assumed that all revenues from ETR were compensated for by reductions in income tax. The only special industry exemption to take into account in Finland was that for very large firms, 10-12 in number, mainly in the paper and pulp industry. However, the scale of the exemption was up to 85 per cent and this is a very large sector in Finland, so an effort was made to include this. According to the Finnish Forestries Industries Federation, the five largest firms made up 85 per cent of turnover in the sector. Therefore it was assumed that the ten largest firms made up 90 per cent of the sector and energy use. Given these assumptions, it seemed reasonable to ignore the relatively small threshold of €50,000 below which all tax is paid, and simply reduce payments from this sector by 100 * 0.85 * 0.9 per cent. Consequently, the pulp and paper sector only paid 23.5 per cent of the tax.

7.3.4.5 THE NETHERLANDS

Much of the processing for the tax rates in The Netherlands was fairly simple in nature, with an average tax rate calculated for heating oil from gas oil and kerosene. Excise duties were not counted as part of the ETR, and any taxes that existed before 1998 were not counted as part of the modelled ETR.

There were major difficulties in estimating industry tax rates and revenues for gas and electricity, however. Energy tax rates for these fuels in The Netherlands are dependent on the size of user (in terms of fuel consumption); in the case of households, it was assumed that all users fell into the smallest category, but this was not a valid assumption for industry. After extensive searching, no relevant data were found for firm size in terms of energy use; while it would have been possible to use the closest data (firm size based on employment or turnover), there was no guarantee that this would have been any more accurate than using a single estimate (and this may also have introduced bias between sectors), so a single category was chosen. This was 50,000-10m kWh of electricity and 170,000-1m cubic metres of gas. In the scenarios with no exemptions, the highest tax rates were used.

A further complication in processing The Netherlands data was that the revenues were disaggregated into just two categories: energy tax and other environmental taxes. As there was no specific mention of exemptions on fuels with the simpler taxes (coal and oil), these revenues were assumed to be correct, and the more complex systems (gas and electricity) were scaled so that the revenues in E3ME matched the published total.

As there was no specific treatment of revenue recycling in the Dutch ETR, it was assumed that the alternative was higher direct income taxes.

7.3.4.6 SWEDEN

The approach for Sweden was very different from the other countries. Statistics Sweden publishes a very detailed set of revenues from environmental taxes, disaggregated by NACE two-digit sector. It was decided to make use of these data, rather than rely on E3ME to estimate the sectoral revenues. This means that a separate rate of exemption was available for each industry. This does assume, however, that the industry exemptions are independent of the fuel mix to that industry—that is, the same exemptions will apply to coal, gas, and renewable energy (although this of course does not mean that the actual tax rates do not vary by fuel type).

As a result of this, the tax revenue data for Sweden were stored as a 19 x 11 matrix (19 fuel users and 11 years) rather than a 12 x 11 matrix (the 12 CT categories and 11 years). E3ME required specific adjustments to cope with this.

Finally, the time series for Swedish tax revenues were extrapolated to include 2003-4. This was done by assuming a linear relationship between the tax revenues and the given tax rates multiplied by projected fuel use.

With no further information, it was assumed that all revenues from ETR were recycled in the form of reduced income taxes.

7.3.4.7 THE UNITED KINGDOM

The climate change levy (CCL) rates were easily obtainable in the UK, but the revenues from the tax are only available as an aggregate for the UK. Individual industries that do not pay the CCL were exempted from the tax during the modelling stage (based on CE's fuel user classification), but the assumption was that exemptions were equal across fuels, so the revenues were allocated to fuels in line with total consumption of that fuel.

By using the data for revenues, the negotiated agreements were taken into account. The scenarios principally looked at the price effects of the CCL and not energy savings made in response to the negotiated agreements; however, the announcement effect on energy demand by other final users was taken into account. Analysis suggests these negotiated agreement effects are not insignificant, so results published here may be understating the overall drop in fuel demand and emissions resulting from the CCL.

As the CCL is a completely new tax in the UK, there was no issue about what counted as part of the 1990s ETR and what was already in place. Revenue recycling was assumed to have occurred completely through the effects of reducing employers' social security contributions.

7.3.4.8 SLOVENIA

Although the CO2 tax in Slovenia was not, strictly speaking, part of an ETR, it was included in the baseline scenario to give an example of environmental taxation in the new member states. For the purpose of the modelling, it was assumed that the revenues were recycled through reductions in direct income taxes.

It was very difficult to define the CO2 tax in Slovenia, with environmental taxes often being bundled with other taxes and different data sources giving conflicting stories. Following consultation, it was decided that only a tax on natural gas consumption should be included. With no data for revenues, it was necessary to assume zero exemptions in all the scenarios, and to estimate revenues as tax rates multiplied by fuel use.

The CO2 tax was not applied to the power generation sector, so it was excluded in the scenarios.

7.3.5 Revenue recycling methods

Table 7.3 illustrates the mechanisms used for

7.3.6 describes these methods in more detail results in each case.

7.3.6 Key assumptions made in the study 7.3.6.1 MODELLING ASSUMPTIONS

Unless otherwise stated, all the modelling follows the same assumptions as the E3ME model. These are documented in the model manual, which is available online at <http://www.camecon-e3memanual.com/>. In addition, the following assumptions were made:

• All of the taxes are revenue-neutral.

Although there are cases where the ETRs are not designed or intended to be revenue neutral, this was imposed in the modelling so that the results indicate the effects of a shift in the tax burden rather than an overall increase or decrease in the tax burden. In cases where the data did not recycling revenue. Section and outlines the expected

Table 7.3. Revenue recycling by country (million C)

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Denmark

Income tax

0

0

0

0

0

0

0

0

0

0

0

Social security contributions

271

615

943

1,032

1,311

1,702

1 , 820

1 ,898

2,044

2,134

2,140

Investment

0

0

8

13

17

28

26

13

0

0

0

Germany

Income tax

0

0

0

0

0

1 ,952

4, 002

5,698

6,951

9,009

9,181

Social security contributions

0

0

0

0

0

1,952

4, 002

5,698

6,951

9,009

9,181

Investment

0

0

0

0

0

201

197

304

183

182

204

Finland

Income tax

0

0

0

373

614

700

685

721

722

895

894

Social security contributions

0

0

0

0

0

0

0

0

0

0

0

Investment

0

0

0

0

0

0

0

0

0

0

0

Sweden

Income tax

-124

-50

349

657

1 ,550

1,570

1,852

1,741

1 ,992

2,395

2,585

Social security contributions

0

0

0

0

0

0

0

0

0

0

0

Investment

0

0

0

0

0

0

0

0

0

0

0

United Kingdom

Income Tax

0

0

0

0

0

0

0

0

0

0

0

Social security contributions

0

0

0

0

0

0

0

540

1 ,372

1,134

1 ,201

Investment

0

0

0

0

0

0

0

0

0

0

0

Slovenia

Income tax

0

0

0

0

0

16

43

57

46

45

45

Social security contributions

0

0

0

0

0

0

0

0

0

0

0

Investment

0

0

0

0

0

0

0

0

0

0

0

Source: Cambridge Econometrics.

Source: Cambridge Econometrics.

Table 7.4. ETR as a percentage of GDP, 2004

DK

DE

NL

FI

SW

UK

SI

ETR Cm

2,140

18,547

2,287

894

2,585

1,200

45

GDP Cm

197,222

2,207,200

489,854

151,935

281,124

1,733,603

26,232

ETR as a % GDP

1.08

0.84

0.47

0.59

0.92

0.07

0.17

Source: Cambridge Econometrics.

Source: Cambridge Econometrics.

support this, shares were used to scale the revenue recycling to match the tax revenues. In cases where there was no clear method of revenue recycling (Finland and Sweden), it was assumed that environmental taxes were an alternative to higher direct income taxes.

• Tax rates in the non-ETR countries and economic activity outside the EU were assumed to remain constant in the scenarios.

7.3.6.2 DATA ASSUMPTIONS

This section summarizes the main assumptions made during the data processing in order to get a complete data set, and reflects our attempts to make best use of the information available.

Where detailed tax revenues were missing, typically the aggregates were shared out using shares of (fuel tax * fuel use), assuming exemptions were similar across fuels or industry. If the literature suggested that there were no exemptions in a particular group, then this total was entered into the data and the remainder of the aggregate tax receipts shared out. Where time series did not cover all of the period 19942004, linear interpolation or extrapolation based on fuel use was used to estimate missing values. Tax rates were assumed to remain constant when no information was available. Tax rates were assumed to remain constant in real terms over the forecast period. Table 7.4 shows that ETR as a percentage of GDP in 2004 was less than 1.1 per cent for all the ETR countries. There are noticeable differences between the ETR countries; in the UK, the ETR accounts for just 0.07 per cent of GDP, compared to 1.08 per cent in Denmark, 0.92 per cent in Sweden, and 0.84 per cent in Germany.

Was this article helpful?

0 0
Going Green For More Cash

Going Green For More Cash

Stop Wasting Resources And Money And Finnally Learn Easy Ideas For Recycling Even If You’ve Tried Everything Before! I Easily Found Easy Solutions For  Recycling Instead Of Buying New And Started Enjoying Savings As Well As Helping The Earth And I'll Show You How YOU Can, Too! Are you sick to death of living with the fact that you feel like you are wasting resources and money?

Get My Free Ebook


Post a comment