3. Tax Revenue Analysis

Historically, tax revenues have been lower and more volatile in Kazakhstan than in comparison to the average for OECD countries. Kazakhstan has undergone considerable economic change over the past two decades. Figure 3.1 shows that the tax-to-GDP ratio in Kazakhstan has been low, volatile and more sensitive to economic cycles. The broad trend over the period has been that the tax-to-GDP ratio rose gradually between 1998 and 2008 and declined gradually thereafter to 2017. GDP declined in 2009 following the global financial crisis and then again 2014 in part due to falling oil prices. The decline in the tax-to-GDP ratio from a high of 27.6% in 2008 to 16.4% in 2017 has also occurred in part because economic growth rose more quickly than tax revenues.

The tax-to-GDP ratio in Kazakhstan is similar to the average of the Commonwealth of Independent States (CIS)1. Following the decline in the tax-to-GDP ratio since 2008, Kazakhstan has a tax-to-GDP ratio of 16.4% in 2017, which is similar to the average of CIS countries (Figure 3.2). For example, it is higher than Azerbaijan (13%), lower than the Republic of Armenia (23%) and similar to Russia (17%). It is also broadly similar to selected Asian and Pacific economies, such as Indonesia (12%), Thailand (18%), the Philippines (18%) and Malaysia (14%) and Singapore (14%). However, the tax-to-GDP ratio in Kazakhstan is low compared to the OECD average (34.2%) and resource-rich OECD countries such as Norway (39%), Canada (33%) and Australia (29%).

The tax mix is concentrated on VAT and CIT with a lower share from PIT and SSCs.2 Taxes on company profits as a share of GDP are high at 5.1% compared to only 3.0% in the OECD. Oil companies contributed to about one-third (35%) of all CIT revenue in 2017. Taxes on goods and services as a share of GDP are 8.5%, which is lower than the OECD average of 10.9%. However, as a share of total tax revenues, taxes on goods and services are relatively high, in comparison to the OECD average. Taxes on the income of individuals are low at 1.4%, compared to 8.3% in the OECD. Similarly, SSC funds are low and represent 0.4% of GDP compared to 9.1% in the OECD.

Kazakhstan’s tax mix is less diversified than in resource-rich OECD countries. As mentioned previously, Kazakhstan aspires to become one of the top 30 global economies by 2050. Figure 3.4 shows developments in taxes for Kazakhstan and a selection of resource-rich OECD countries and emerging economies between 2000 and 2017. A diversified tax mix reduces the potential impact of a decline in any one tax. In Kazakhstan, 81% % of tax revenues come from two taxes – taxes on goods and services and taxes on companies – based on OECD revenue statistics data. In Norway, Canada and Mexico, the tax revenues from the same two taxes are lower at 43%, 35% and 58% respectively. The reduced degree of tax concentration among these countries provides them with greater protection from potential shocks.

The lack of diversification in the tax mix have led to larger declines in tax revenues than in resource-rich OECD countries and emerging economies. While PIT and SSCs have both been relatively stable in Kazakhstan and Norway in recent years, an important difference arises in the size of the contribution of these taxes to total tax revenues. Combined, PIT and SSCs represent over 20% of GDP in Norway in 2017 but less than 2% in Kazakhstan. When the international oil price declined between 2014 and 2016, this combination of a proportionately high (relative to total taxes) and stable tax base provided Norway with additional protection for tax revenues. As a result, during the period 2012 to 2017, total tax revenues fell by a third (from 23.9% of GDP to 16.4%) in Kazakhstan compared to just 3% in Norway (from 41.5% to 38.8%).

CIT and export duty tax (on oil) revenues depend on the oil sector. Figure 3.5 shows the contribution of oil production to different taxes based on data provided by the authorities. Overall, 42% of total tax revenues are related to the oil sector in 20183. This has fallen from a recent peak of 54% in 2011. In the case of PIT revenues, 9% is related to the oil sector and the figure has remained relatively stable over time. For revenues from the CIT, 45% is related to the oil sector and the trend has followed overall tax revenues trends over the period. In the case of VAT revenues, the proportion is 11% having increased from negative values (one possible explanation is because of large VAT refunds for VAT paid on inputs). For export tax revenues, virtually all revenues are derived from the oil sector (97.9%). Export custom duties relate to taxes on crude oil and oil products (these taxes are levied on the basis of the quantity exported). Taken together, CIT and export taxes represent over 90% of all taxes from the oil sector.

Oil and non-oil tax revenues are sensitive to changes in the oil price. The international crude oil price has a strong positive correlation with overall tax revenues in Kazakhstan over time, particularly with tax revenues from the oil sector (Figure 3.6). For example, between 2014 and 2016, the oil price fell by 48% and 11% and oil tax revenues fell similarly by 48% and 23% respectively. When the oil price recovered in 2017, so did oil tax revenues, albeit more quickly. The impact of the oil price on the economy and total tax revenues operates through various channels including through increased company profits which are subject to tax (for more details on company taxation see Chapter 5). The data also show that the oil price has a positive correlation with taxes from the non-oil sector, which reflects the integrated nature of oil in the Kazakhstan economy.

The volatility of CIT revenues is high. Among the selected resource-rich countries, Kazakhstan and Norway have had similarly high degrees of CIT revenue volatility, measured based on the standard deviation in revenues over the 2000 to 2017 period (Table 3.1). The standard deviation was 2.7 and 2.5 in each country respectively over the 18-year period. The trend can also be observed in the tax-to-GDP ratios: between 2012 and 2017, for example, CIT revenues fell from 7.7% to 4.5% in Kazakhstan and from 10.3% to 4.6% in Norway, which is a broadly similar relative decline. By comparison, Chile has had a more moderate CIT revenue volatility and Brazil and Canada have had low CIT revenue volatility.

Taxes on goods and services are more volatile than in other resource-rich countries. Between 2014 and 2015 when the international oil price fell, taxes on goods and services fell sharply in Kazakhstan by 33%; while in Norway they rose modestly by 3.0%. More generally, goods and services appears to be more volatile in Kazakhstan than in other resource-rich countries, based on the standard deviation over the 18-year period from 2000 to 2017 (Table 3.1).

VAT represents a relatively stable form of tax that is linked to the overall level of domestic consumption in many OECD countries, but this does not seem to be the case in Kazakhstan. To take one example, in Norway VAT and consumption have been closely correlated over the past decade Figure 3.7. In Kazakhstan, however, VAT is correlated with the general level of consumption but less so than in other resource-rich countries such as in Norway. One explanation for this could be the rise in the importance of imports in the VAT base that has occurred in recent decades. For example, VAT on imports as share of total VAT increased from 26% in 1998 to 78% in 2008 to 61% in 2017.

Higher tax revenues are needed to meet Kazakhstan’s expenditure goals and to reduce the non-oil deficit. (see Infographic 2.1). Raising more tax revenues would allow the country to meet its revenue targets and expenditure goals and to reduce the non-oil deficit. The authorities have set an ambitious goal to increase tax collection from 18 to 25 percent of GDP by 2025 (IMF, 2020[1]). The authorities also have a wide range of ambitious expenditure goals including supporting low-income households, promoting SMEs and enhancing the quality of healthcare. In addition, Kazakhstan presently runs a budget deficit and a non-oil budget deficit (NOD). In general terms, the idea of NOD is to strip out oil revenues from the budget to provide a more realistic measure of the budget deficit in the absence of oil revenues. The authorities define the NOD as the difference between budget revenues (with the exception of loan receipts, transfers from the National Fund and export customs on crude oil) and expenditures (with the exception of repayment of loans). Reducing the NOD has been identified by government as a reform priority.

A further reason to raise tax revenues is that government debt has been rising in recent years albeit from a low base. Furthermore, over the long-term, the low-carbon transition and technological advancements in fossil fuel extraction are likely to generate downward pressure on fossil fuel demand, prices and the associated tax revenues (Botta, 2020[2]). To prepare for decarbonisation of the world economy over the longer-term, Kazakhstan has recognised that it should gradually raise more tax revenues from non-resource related sources to finance its budget. The resource revenues should provide additional time for a gradual transition towards raising higher non-oil related tax revenues in ways that do not harm growth, respect fairness and is aligned with the capacity of the tax administration to operate a modern and well-designed tax system. A broader set of macroeconomic risks exist including the potential for increases in international interest rates, oil price shocks and trade wars.

Kazakhstan needs to strike a balance between consuming resource-revenue today versus investing and consuming the revenues in the future. Resource-rich countries such as Kazakhstan face the opportunities and challenges of managing oil, gas, and mineral revenues that are volatile, finite and uncertain (Botta, 2020[2]). As the world gradually decarbonises over the long-term, Kazakhstan can decide how much resource-revenue to consume and invest now through the budget and how much to save for the future through the National Fund. The analysis in Infographic 2.2 provides an indicative snapshot in 2017 of the flows of oil and non-oil tax and non-tax revenues in the Kazakhstan fiscal system. Some caveats are worth noting. The GDP shares shown in Infographic 2.2 are derived based on different sources (including OECD revenue statistics for the budget and most tax revenues) so they cannot necessarily be combined. Similarly, the GDP shares are for the year 2017 (unless otherwise stated) and are not necessarily representative of other years, partly because of the volatility of the economic activity and profitability of the oil sector (for example, investment income under management of the National Fund and transfers to budget can change significant in a given year).

Resource revenues currently fuel the National Fund and the budget. Notwithstanding the illustrative nature of the analysis in Infographic 2.2, it highlights a number of points. The tax and non-tax revenues generated by the oil sector in 2017 are significant (5.5% of GDP). Most tax and non-tax oil revenues flow to the National Fund (3.7% of GDP) but some go directly to the budget (1.7% of GDP). Non-oil tax revenues go directly to the budget (11.4% of GDP) along with non-oil non-tax revenues (1.1% of GDP). Kazakhstan’s oil wealth is managed through its National Fund. The assets of the National Fund are significant at 33% of GDP in 2017. The assets are accumulated in a number of ways. They include taxes from oil sector organisations including CIT, alternative subsoil use tax, mineral extraction tax, bonuses, export rental tax, excess profit tax. They also includes other revenues from operations carried by organisations of the oil sector, including for example violations of the terms of oil contracts in addition to other proceeds (EITI, 2018[3]). Assets also includes investment income from the management of the fund (2.7% of GDP). The National Fund annually allocates transfers to the budget (8.3% of GDP in 2017) but these do not necessarily correspond to the revenues flowing into the National Fund. Some of these transfers can be made in accordance with the decision of the President (referred to as targeted transfers). Budget revenues are estimated at 22.5% of GDP, comprised of oil tax revenues including export custom duty (1.7%), non-oil tax revenues (11.4%), non-oil non-tax revenues (1.1%) and transfers from the National Fund (8.3%) (for the year 2017, when transfers were high relative to previous years).

Tax revenues flow to the budget and the National Fund. Tax revenues from the oil sector in Kazakhstan go to the National Fund with the exception of export customs duty on crude oil. Therefore, oil tax revenues equal total tax revenues that go to the National Fund plus export customs duty on crude oil that go to the budget. The corollary is that non-oil tax revenues are total tax revenues less oil tax revenues. Table 3.2 shows the components of oil and non-oil tax revenues in Kazakhstan between 2010 and 2017 using OECD revenue statistics data. Export customs duty on crude oil can be approximated using OECD revenue statistics with taxes on exports excluding those that go to the National Fund.4

An OECD definition of the NOD can be developed. A number of methodological steps are taken to estimate a new definition of the NOD to 2025 for the purposes of this report. ) First, using OECD revenue statistics data, non-oil tax revenues are calculated as the amount of tax revenues that do not flow to the National Fund but directly to the general government budget (see Table 3.2). Second, non-oil non-tax revenue is added to non-oil tax revenue to give total non-oil revenues. It is worth noting that the definition of what constitutes tax and non-tax revenues differ in Kazakhstan and the OECD revenue statistics5. Third, ‘taxes on exports excluding those to the National Fund’ are also deducted from the non-oil tax revenue estimate as these relate to export tax duties on crude oil from the oil sector. Finally, state budget expenses based on State Revenue Committee data are deducted to give an estimate of the non-oil deficit. Based on these steps, the definition of the NOD can be expressed as follows:

NOD = tax revenues to budget + non-tax revenues to budget – taxes on export duties – expenditures 6

The NOD is presented in Table 3.3 for years up to 2017, using actual data and forecasts the NOD for the period 2018-2025. After 2017 (the latest year for which there is OECD revenue statistics data), non-oil tax revenues, GDP and expenditures are grown-forward annually based on international data and forecasts (IMF, 2020[1]).

The NOD could range from 4 – 7% by 2025. According to the baseline analysis, the NOD is 5.4% of GDP in 2018, 6.2% of GDP in 2020 and could be 3.5% by 2025.7 However, the forecast growth in expenditures (based on IMF data) are low between 2018 and 2025, averaging 0.2% over the period, and resulting in a gradual decrease over time of expenditure as a % of GDP. For that reason, Figure 1.1 extends the analysis from Table 3.3 and adds a simulation analysis to show the NOD if expenditure growth were 2 and 3 percentage points higher annually than the baseline growth rates. On this basis, the NOD could range from 4 – 7% of GDP by 2025 (estimates are rounded).

The government NOD forecast sits in between this range, albeit the definition of NOD is different.8 The authorities forecast the NOD at 7.4% of GDP in 2018, 7.0% in 2020, 6.5% in 2022 and 6.0% in 2025.9 This NOD estimate is in the range of previous international estimates ranging from 5% to 8% for 2025 based on various fiscal adjustment strategies (World Bank, 2017[5]). However, these forecasts may depend on relatively optimistic underlying forecasts of non-oil revenue, expenditures and GDP (higher GDP forecasts reduce the NOD).

Reducing the NOD would require raising significantly more tax revenues. The tax-to-GDP ratio is 16.4% in 2017; the figure includes the SSCs that are paid to separate social funds. On the basis of the previous simulation analysis, to eliminate fully its NOD, Kazakhstan will have to raise its tax-to-GDP ratio by 4 - 7 percentage points. This financing need for the NOD is the starting basis for the analysis below, which examines which taxes it might be raised from. Research conducted by the Economic Research Institute in Kazakhstan reaches a similar result showing that the tax-to-GDP ratio could be increased by 4 – 5 percentage points within 10 years (from the level of 2018) with limited negative effects on economic growth and tax collection (Alpysbayeva, Kenzhebulat and Karashulakov, 2019[6]).

Financing needs could be met by gradually moving some taxes that have the potential for generating more revenues and are less dependent on revenues from the resource-sector. Kazakhstan’s tax base is concentrated in taxes on goods and services (8.4% of GDP) and taxes on companies (4.5%) and to a lesser extent taxes on individuals (1.4%), SSCs (0.5%) and property taxes (0.5%) (see Chpater 3). To support its financing needs and simultaneously increase the diversity of the tax base, Kazakhstan could support the financing of the NOD by raising tax revenues from taxes that ‘underperform’ relative to the OECD average and which are less exposed to the resource-sector.

There are several ways in which taxes could be raised to finance the NOD. Tax revenue increases could come from a combination of PIT, SSCs and property taxes along with a modest increase in taxes on goods and services. Table 3.4 shows the additional taxes as a share of GDP that would be required to cover a NOD of 6% of GDP for example, which lies between the 4 – 7% of GDP shown in the previous simulation analysis. For example, taxes on goods and services as a share of GDP are 8.4% in 2017. According to the analysis shown in Table 3.4 an additional 0.6% in taxes on goods and services would be needed to finance the NOD by 2025, along with the other increases in tax revenues shown. This analysis provides a framework for considering which taxes, and how much revenue from each tax, might need to be raised. A range of revenue raising alternatives are possible across the different taxes. As part of the recommendations in this report, there is additional scope for raising tax revenues from PIT including tax on personal capital income, SSCs, property taxes and VAT. For example, one option could be increases in taxes on individuals, SSCs and property taxes while modestly increasing taxes on goods and services) and maintaining current company tax levels. The scheduled increase in SSCs may contribute to lowering the NOD, but only to the extent to which the higher SSCs will finance social spending that already takes place. If the scheduled increase in employee and employer SSCs will finance additional social spending, the reform will leave the current NOD intact, which then will have to be financed through other tax increases. Maintaining revenues of company taxes constant over time would, in an environment of decreasing oil prices because of decarbonisation of the world economy, imply a gradual increase in company taxes paid by the non-oil sector in Kazakhstan. While reaching the OECD average should not be thought of as a goal in itself, it provides a framework for identifying where there may be potential to generate more tax revenues from less volatile tax revenue sources.

References

[6] Alpysbayeva, S., M. Kenzhebulat and G. Karashulakov (2019), “The Potential for Increasing Tax Revenues Amidst Reducing Fiscal Inconsistencies Zone(based on Kazakhstan’s economy researches)”, HSE Economic Journal, Vol. 23/3, pp. 365-383, https://ideas.repec.org/a/hig/ecohse/201932.html (accessed on 6 August 2020).

[2] Botta, E. (2020), The fiscal implications of the low-carbon transition, http://www.oecd.org/greengrowth (accessed on 14 February 2020).

[7] Brys, B. et al. (2016), “Tax Design for Inclusive Economic Growth”, OECD Taxation Working Papers, No. 26, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jlv74ggk0g7-en.

[3] EITI (2018), Extractive industry transparency initative.

[1] IMF (2020), Republic of Kazakhstan : 2019 Article IV Consultation-Press Release; and Staff Report, https://www.imf.org/en/Publications/CR/Issues/2020/01/29/Republic-of-Kazakhstan-2019-Article-IV-Consultation-Press-Release-and-Staff-Report-49002?cid=em-COM-123-41052 (accessed on 11 February 2020).

[4] OECD (2020), Global Revenue Statistics Database, https://stats.oecd.org/Index.aspx?DataSetCode=RS_GBL (accessed on 4 August 2020).

[5] World Bank (2017), Enhancing the fiscal framework to support economic transformation, https://openknowledge.worldbank.org/bitstream/handle/10986/28939/121677.pdf?sequence=5&isAllowed=y (accessed on 29 August 2019).

Notes

← 1. The CIS include Armenia, Azerbaijan, Belarus, Kazakhstan, Kyrgyzstan, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan.

← 2. It should be noted that the definition of what constitutes tax and non-tax revenues differ in Kazakhstan and the OECD revenue statistics. The share of the Republic of Kazakhstan under production sharing contracts of oil companies, the bonuses of oil and non-oil sector companies, the levy for the use of the radio-frequency spectrum, the payment to compensate for historic costs as well as certain other items are classified as non-tax revenues according to the OECD Interpretative Guide, but are considered as tax revenues in Kazakhstan.

← 3. Note that the OECD global revenue statistics database show that oil sector tax revenues are 33% of total tax revenues in 2018.

← 4. In 2017 for example, taxes on exports excluding those to the National Fund (OECD revenue statistics) and exports custom duty on crude oil (State Revenue Committee of Kazakhstan) were KZT 904 476 and KZT 854 530 respectively, a difference of only of 5.8%.

← 5. The share of the Republic of Kazakhstan under production sharing contracts of oil companies, the bonuses of oil and non-oil sector companies, the levy for the use of the radio-frequency spectrum, the payment to compensate for historic costs as well as certain other items are classified as non-tax revenues according to the OECD revenue statistics Interpretative Guide. The classification used by Kazakhstan does consider these revenues as tax revenues. This report follows the tax definition and classification from OECD revenue statistics.

← 6. Under this definition, tax revenues to the budget exclude taxes paid by the oil sector (but include export duties on crude oil) and non-tax revenues paid to the general budget exclude non-taxes paid by the oil sector.

← 7. The decrease in the NOD in 2018 is largely due to the reduction in current expenditures in that year.

← 8. The NOD is defined by the authorities in Kazakhstan as the difference between budget revenues (with the exception of loan receipts, transfers from the National Oil Fund and export customs on crude oil) and expenditures (with the exception of repayment of loans). These forecasts are according to the Presidential degree in 2016.

← 9. Forecast of Department of Macroeconomic Analysis and Forecasting. Ministry of National Economy (2019).

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

https://doi.org/10.1787/872d016c-en

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.