REVENUE PRODUCTIVITY OF THE TAX SYSTEM IN LESOTHO

Nthabiseng J. Koatsa and Mamello A. Nchake

Department of Economics
National University of Lesotho

Abstract

Fiscal policy is the only macroeconomic policy tool at the disposal of Lesotho government
because the country gave up its monetary policy, trade policy and exchange rate policy upon
joining SACU and CMA. The study evaluates the revenue productivity of Lesotho’s overall tax
system as well as of the major components of tax revenue, income tax and value added tax, using
annual time-series data for the period 1992-2015. The buoyancy of the tax system is measured
using both the traditional regression approach as well as the dummy variable approach. The
results indicate that overall, the Lesotho tax system and its major components are elastic and
buoyant as the coefficient of log (GDP) is statistically significant and greater than unity in all the
regressions. The results also indicate that delays in the implementation of tax policies announced
in the budget speeches negatively affect tax revenue. The introduction of the Lesotho Revenue
Authority (LRA) in 2003 seems to have marginally improved the efficiency of the tax system in
general. Other tax reforms do not seem to have significantly improved on the revenue
productivity of the system as they were focused more on equity. One conclusion from the study
is that the tax revenue in Lesotho grows due to increases in income rather than improvement in
government effort in tax collection. Although the equity focus of the tax policy is commendable,
there is a need to focus on revenue productivity as well such that future reforms are geared
towards improving the efficiency in tax collection.

1
 1. Introduction
Notwithstanding the considerable debate concerning which services should be provided by the
state, there is a general consensus that raising the level of government expenditure, especially for
public investment in key areas of the economy, tends to be an important ingredient in the
development process. The importance of taxation lies in its ability to generate sufficient revenue
to meet expanding requirements of the public sector. Taxation is also used to regulate the
economy by influencing vital economic variables, employment, income and wealth distribution,
and in some cases the pattern of consumption.

Tax revenue mobilization plays an important role especially in developing countries where it is a
major source of domestic revenue for financing critical public expenditures needed for growth.
In most cases, it is deemed to be a better source of revenue mobilization than other sources such
as deficit financing and money creation. However, most developing countries find it difficult to
raise enough tax revenues1.

Figure 1 shows that the ratio of tax revenue to GDP in Lesotho between 1992 and 2002 was
constant and averaged 13%. Following the introduction of the Lesotho Revenue Authority in
2003, there is an upward trend in the ratio.

Figure 1: Ratio of Tax Revenue to GDP, 1992-2015

Source: Central Bank of Lesotho

1
The reasons for this vary between countries but it can be concluded that in developing countries tax policy
focuses on what is possible rather than the pursuit of the optimal.

2
The ability of a tax system to generate revenue has a bearing on the services that can be provided
by a government. Hence, for taxation to be an effective resource mobilisation tool the tax system
needs to be productive or elastic. The productivity of a tax system may be measured using two
concepts: tax elasticity, and tax buoyancy. Tax revenue may change due to a number of factors
such as, changes in tax rates, tax base, as well as efficiency of tax assessment and collection. Tax
elasticity measures the responsiveness of tax revenue to changes in income while tax buoyancy
considers the responsiveness of tax revenue to changes in income as well as to changes in
discretionary measures such as tax rates and tax bases. It is desirable that the income elasticity
and buoyancy of a tax system should be equal or greater than unity (Bonga et al, 2015). The
distinction between tax elasticity and buoyancy is valuable when evaluating whether future
revenues will sufficiently meet the resource needs without changing the rates or bases of existing
taxes (Timsina, 2006)

Many developing countries still face difficulties in raising adequate tax revenue with many of
these countries suffering from an over-dependence on a small number of sources of tax revenue
such as export taxes on mineral products which are vulnerable to external factors,. In an attempt
to curb this problem, many developing countries undertook tax reforms during the 1980s.
However, according to Osoro, (1993) most of those reforms were on tax structure with the
general aim of revenue adequacy, equity and fairness, simplicity and economic efficiency, rather
than on tax administration. Since independence, Lesotho’s tax system has also undergone several
tax reforms.

The efficiency of the of the taxation system is very important in Lesotho, because as a member
of CMA and SACU, the country has no control over its monetary policy and depends heavily on
SACU revenues which are exogenously determined (see Figure 2) hence fiscal policy is the only
macroeconomic tool for influencing the economy. Recent declines and uncertainties in SACU
receipts coupled with escalating public expenditures, mostly driven by the high wage bill (see
Figure 3), have plagued the fiscal stance of the nation and have necessitated an interrogation of
alternative means of internal revenue mobilization.

3
Figure 2: Government Revenue, SACU Receipts and Tax Revenue (1992-2015)

Source: Central Bank of Lesotho

Figure 3: Share of the Wage Bill in Lesotho (1992-2015)

Source: World Development Indicators

Figure 4 which presents information on the components of Lesotho’s tax system shows that over
half of the tax revenue is raised from income tax and about a third from VAT.

4
Figure 4: Tax revenue performance (period averages): 1992-2016

Source: Central Bank of Lesotho

Given the importance of taxation to the Lesotho economy, studies on the productivity of the tax
system are essential. Based on the foregoing discussion, the main objective of this study is to
measure the elasticity and buoyancy of Lesotho’s overall tax system to ascertain whether it is
revenue productive. Short-run and long-run elasticities and buoyancy will be estimated for tax
revenue and its main components of income tax, value added tax (VAT) for the period 1992 to
2016.

2. OVERVIEW OF TAX REFORM IN LESOTHO

Tax reform is necessitated by a deficiency of the existing tax system to meet the stated goals. It
usually addresses four main concerns: why it is done, when it should be done, where it should be
done and how it should be done. Lesotho’s tax reform dates back to shortly after independence,
in the late 1960s. However, the major tax reforms commenced in the 1990s. Most of the reforms
were undertaken on the income tax structure with the main objective of collecting more revenue
in order to reduce financial imbalances in the economy. Other objectives that are addressed by
these reforms include simplicity as well as equity. These changes resulted in reduced number of
tax brackets and increased chargeable income.

Tax reform shortly after independence

In 1968 the basic and graded tax was replaced by a graduated personal tax. The minimum rate of
tax was raised by approximately 35%, to R4.50 per annum. The graduated tax had two tiers, with

5
the minimum rate of R4.50 applied on incomes up to R500 per annum, and the tax increased
cumulatively by 50% per completed R100 of income on any incomes in excess of R500. An
individual paid whichever was greater between graduated personal tax and income tax.
Tax reform in the 1980s
In 1982, the Department of Sales Tax was established and a general sales tax (GST) was
introduced at the rate of 5%, and later increased to 6% in 1984. Also tax on alcohol was
increased from 17.5% to 22% with effect from 1 April 1988. All these reforms were aimed at
revenue generation.
Tax reform in the 1990s
In the 1990s, several reforms were undertaken mostly on income tax, with just one reform on
sales tax. However these reforms were biased towards the structure of the tax structure, and
ignored tax administration. In 1990, abatements relating to dependents were abolished and
medical allowances were terminated in order to increase revenue and to reduce tax evasion and
avoidance (Selialia, 1993). Abatements are non-taxable incomes allowed to tax payers based on
marital status, with married allowances being greater than single allowances. During the period
1990-1993, every time there was a change in the tax rate structure, the abatements allowances
were also increased.

In 1991, sales tax rate was increased to 10%. During the same period, the income tax system had
five brackets, with the lowest bracket taxed at 15% and the highest bracket taxed at 53%, with an
average marginal tax rate of 35%. In 1992, the highest tax bracket was reduced from 53% to
48% resulting in the average marginal tax rate falling to 33%.

In 1993, the tax brackets were reduced from five to three, with the lowest income taxed at 25%
and the highest rate at 40%. Thus, the reduction in the number of tax brackets was accompanied
by an increase in the lowest tax rate from 15% to 25%. In 1996, the three-band structure of 25%,
35% and 40% was replaced with a two-band structure of 25% and 35%, with the lowest income
bracket charged at 25% and the balance of chargeable income taxed at 35%. During the same
period, the abatements system was replaced with single personal tax credit system. The tax credit
can be defined as a reduction in tax liability or a tax saving or a negative tax. The difference
between the tax credit and abatements is that the former is applicable to all tax payers at a fixed
amount regardless of income level and marital status. The tax credit was set at M2 640 per
annum per taxpayer. The introduction of the tax credit meant that a taxpayer had to earn more
6
than M10 560 per annum to be eligible to pay income tax. Corporate tax rates were reduced from
40% to 35% with the aim of encouraging businesses as well as bringing corporate tax in line
with regional tax rates.

Tax reform in the 2000s

In an attempt to strengthen tax administration, the government of Lesotho established Lesotho
Revenue Authority (LRA) in 2003 and introduced Value Added Tax (VAT) to General Sales Tax
(GST). The purpose of the VAT was to stop the abuse of tax exemptions certificates and to close
the loopholes that suppliers were using to evade tax. It was hoped that these measures would
widen the tax base and improve efficiency and equity.

In 2006, corporate tax rates were reduced from 35% to 25% with the objective of encouraging
private sector growth, and improving competitiveness with South Africa, with both actions
hoped to increase corporate tax revenue In addition, the tax on manufacturing industries was
lowered from 15% to 10% to attract investors and provide support for manufacturing industries.
An upfront VAT refund scheme to cover all exporters was proposed as well as a zero percent
company tax on income generated from exporting manufactured goods to countries outside the
SACU region.
In 2007 tax credit was increased from M2 911 to M3 500 mainly to help low income groups.
Further income tax reform included a reduction of the lower tax rate from 25% to 22% and the
increase of the threshold from M35 064 to M37 378. The threshold where people start paying tax
was also raised from M11 643 per annum to M14 000. In 2009 personal income tax was linked
to inflation, and the tax credit was increased from M4 500 to M5 000.

In 2012, the tax credit was adjusted to M 5 755 as a response to the difficulties endured by
income earners due to the global economic crisis, implying that income less than M26 160 per
annum attracted no tax. In 2014, in addition to increasing the tax credit and the thresholds by an
inflation rate of 6%, both the lower and the upper personal income tax rates were also reduced
from 22% to 20% and 35% to 30% respectively. The tax credit and the threshold were adjusted
to avoid tax brackets creep. The tax credit was increased to M6 100 implying that the lowest
taxable income was increased to M27 730 while the threshold for higher earners was adjusted to
M51 690. During this same period, other tax reforms that took place include: the so called sin
taxes on liquor and cigarettes; taxation of vehicles based on their gas emission were harmonized

7
at SACU level; the minimum turnover threshold that businesses have to register for VAT was
increased from M500, 000 to M850, 000, providing relief for some small businesses.

3. LITERATURE REVIEW

Tax elasticity and tax buoyancy are measures of the effectiveness of a country’s tax strategy. A
tax buoyancy value of less than one suggests ineffective discretionary changes while a value
greater than one implies that the discretionary policy changes are improving the tax system.
Income elasticity of a tax is a product of the tax-to-base and base-to-income elasticities. Thus the
composition and changes in the tax base as well, as the factors that determine that base are
important determinants of income elasticity of tax. Buoyancies and elasticities can be calculated
for the entire tax regime or for individual components of the tax system. Generally, regressive
and specific taxes tend to contribute to low tax elasticities and buoyancies.

The concept of revenue productivity of a tax in terms of buoyancy and elasticity was first
introduced by Sahota (1961) using the data for India. Sahota and subsequent studies such as
Nambiar and Rao (1972); Rajkumar and Chdambaram (1972); and Howard (1992) found large
variations in tax elasticities and buoyancies in both advanced and developing countries (Fauzia,
2001; Mukarram, 2001; Cotton, 2012). Studies on revenue productivity of tax systems in Africa
by researchers, notably, Osoro (1993), Moyi and Muriithi (2003), Kusi (1998), Chipeta (1998)
and Ariyo (1997) provided inconclusive results because of data constraints. Various models have
been used to estimate the tax buoyancy and elasticity. Some studies used the dummy variable
model, others the proportional adjustment (PA) technique, while others only estimated
elasticities due to lack of sufficient data on changes in tax policies which are needed to compute
buoyancies. A summary of some key empirical studies from Africa is highlighted in Table 1.

8
 Table 1: Summary of the key empirical studies using data from African countries
Author (s) Data Model Estimation Key findings
coverage technique
Osoro (1993) Tanzania, Proportional OLS and Tax reforms did not increase the
1969-1990 adjustment decomposition revenue productivity of the tax
system and tax elasticities are
inelastic for the major taxes and
the general tax system.
Kusi (1998) Ghana, Proportional Tax reform positively contributed
1960-1993 adjustment to
revenue productivity growth and
tax
elasticities are elastic for the
major taxes.
Botlhole and Botswana, Singer dummy vector error Income elastic and buoyant tax
Agiobenebo 1982-2001 variable model correction system. Mineral tax revenue is
(2006) model buoyant and elastic with respect to
(VECM), mining GDP; non-mineral income
tax is buoyant and elastic with
respect to exports; customs and
excise duties are neither buoyant
nor elastic.
Muriithi and Kenya, Proportional Tax reforms had positive impact
Moyi 1973-1999 Adjustment on the overall tax structure and on
(2003) the individual tax handles.
Ehdaie (1990) Malawi Dynamic 3Stage Least Discretionary measures are highly
and simultaneous- Squares significant
Mauritius, equation
1965-1985 econometric
model

9
Kargbo and Sierra Singer dummy Engle-Granger Discretionary tax measures are
Egwaikhide, Leone, variable model Two-Step effective in tax revenue
2012 1977 -2009 cointegration productivity and the general tax
approach system is inelastic.
Tax buoyancy estimates were
more elastic than tax elasticity
estimates. Short-run elasticities are
lower than the static long-run
elasticities.

Bayu, 2015 Ethiopia, Traditional Cointegration Gross, direct and domestic indirect
1974-2010 model approach tax revenues are non-buoyant both
in short run and in the long run.
Foreign trade tax are non-buoyant
in the short run but buoyant in the
long-run
Bonga, Zimbabwe, Traditional Cointegration The general tax system is buoyant
Dhoro- 2000-2013 and Singer approach and elastic.
Gwaendepi, Dummy
Mawire-Van Variable
Strien, 2015 models

4. METHODOLOGY

Data sources and description

The study uses annual data from 1992 to 2015. The data on tax revenue was obtained from the
Central Bank of Lesotho while the data on various measures of GDP was obtained from World
Bank Indicators database.

Conceptual framework

By the traditional definition of a tax system, tax elasticity and buoyancy are determined by the
tax base, tax rate and tax structure. The concept of tax elasticity is therefore defined as the
percentage increase in the tax revenue resulting from endogenous changes in the tax base caused
by a percentage rise in Gross Domestic Product (GDP). The income elasticity of a tax can be
broken down into tax-to-base and base-to-income elasticities, with the elasticity of a tax being
the product of the two components. The composition of the tax base can change significantly
over relatively long periods, especially in a developing country. Singer (1968) introduced the
concept of ‘tax buoyancy’ by incorporating the dummy variable and thereby altering the

10
traditional approach to capture the exogenous influences resulting from tax legislation on the tax
base, tax rate and / or tax structure. This study draws particularly from both approaches to
measure elasticity and buoyancy of the Lesotho tax system for the period 1992-2015.

Model Specification
The empirical analysis in this study starts with the multiplicative functional form of a tax
revenue model represented in equation (1):
(1)

where is the total tax revenue, is the tax base proxied by real GDP, is a constant term and
is the coefficient of income and is the error term. The total tax revenue comprises income
tax, taxes on goods and services (customs and excise duties, VAT/Sales tax), corporate profit tax
and other taxes. The nature of Lesotho’s tax revenue sources is such that SACU revenues
account for the large share of the total government receipts. Our data allows us to estimate the
revenue productivity of a tax system through the buoyancy and the elasticity of a tax system.
The tax buoyancy is estimated using the double log-linear form of equation (1) represented by
equation (2):

(2)

Where; is the log of total tax revenue; is the log of real GDP and the log of real GDP
per capita; is the buoyancy coefficient in year t
Equation (2) is modified based on Ariyo (1997), by including one year lag of GDP to account for
any delays in tax policy implementation in period t that may affect tax base to obtain equation
(3).

(3)

Where; is the previous year’s income; and is the buoyancy coefficient for the previous
year.
One limitation of the preceding approach is that it does not account for changes in tax policy and
in the institutional framework. To overcome this limitation, this study extends the analysis by

11
introducing the dummy variable for each year in which there was an exogenous tax policy
change, as proposed by Singer (1968)2. Introducing these dummy variables gives equation (4).

∑ (4)

Where are the intercept dummy variables taking the value of 1 for the years in which there
were discretionary policy changes and 0 otherwise. are the coefficients of intercept dummies
of which k is the number of dummies for changes in tax policy. is aggregate tax elasticity in
the current period while is the tax elasticity in the previous period. Both coefficients measure
the percentage increases in the tax revenue resulting from the endogenous changes in the base
caused by a one percent rise in GDP.
From equation (4), the log of total tax revenue estimates the productivity (both the buoyancy and
the elasticity) of the tax system in Lesotho.

5. RESULTS

Stationarity Tests

The next step in our analysis is to establish the stationarity of the variables in question prior to
estimating the economic relationships. Table 2 presents the results for the Augmented - Dicky
Fuller (ADF) and Phillips-Perron (PP) Tests for all the variables used in this study. If the
calculated test statistic is greater than the critical values, then we reject the null hypothesis of a
unit root and alternatively if the test statistics is less than the critical values, we do not reject the
null hypothesis.
These results suggest that all the variables are stationary in levels with and without the trend
term. Hence we conclude that all the variables are integrated of order 0, that is, they are I (0).
Hence there is no need to further transform the data.

2
Chand and Wolf (1973), Khan (1973) and Artus (1974) also used the same methodology.

12
 Table 2: Unit root tests (1992-2015)
ADF PP
Intercept Intercept and
VARIABLE Intercept and Trend Intercept Trend
Log(Total Tax) 0.356** -3.426* 0.611** -3.414*
Log (Income Tax) -0.307** -2.865** -0.300** -2.901**
Log (PAYE) -0.142** -2.555** 0.099** -3.677*
Log (VAT) -1.143** -3.707* -1.227** -2.547**
log(GDP) -1.746** -3.100** -2.928** -3.048**
CRITICAL VALUES
1% -3.750 -4.380 -3.750 -4.380
5% -3.000 -3.600 -3.000 -3.600
Note: ** significance level at 1% , * significance level at 5%

The next step was to test for the stationarity of the corresponding residuals in regressing drawing
from the two step cointegration test proposed by Engel and Granger (1987). We estimate Log
(GDP) on Log (Total Tax), Log (Income Tax), Log (PAYE) and Log (VAT) using OLS and test
the stationarity of the corresponding residuals. The test statistics for the residuals are greater than
the critical values at 1%, suggesting that the four equations for total tax, income tax, PAYE and
VAT with nominal GDP are stationary. Thus we conclude that measures of tax revenue have the
long run relationship with GDP.

Basic regression: Tax elasticity

Given the above conclusions, we can proceed to use OLS to measure the tax elasticity of
Lesotho’s tax system by estimating equation (1) and equation (2). The results are presented in
Table 3.
Overall, the Lesotho tax system and its major components (income and VAT) are elastic as the
coefficient of log GDP is statistically significant and is greater than unity in all the regressions.
VAT is the main driver of tax revenue as it is more elastic than PAYE. On the other hand,
income tax which includes PAYE and corporate taxes has higher elasticity than VAT indicating
that the government can collect more revenue from the private sector through the corporate
taxes.
As already mentioned, lags in tax administration due to delays in the implementation of tax
policies announced in budget speeches negatively affect tax revenue. This is particularly
important for income tax which has a highly statistically significant coefficient of lagged GDP.

13
 Table 3: OLS regression results on tax elasticity in Lesotho (1992 -2015)
(1) (2) (3) (4) (5) (6) (7) (8)
VARIABLES

Log (GDP) 1.221*** 1.246*** 1.245*** 1.239*** 1.140*** 1.143*** 1.231*** 1.258***
(0.033) (0.029) (0.019) (0.019) (0.018) (0.023) (0.043) (0.046)
Log (GDP -1) -0.244 -0.975** 0.055 -0.245
(0.405) (0.394) (0.527) (0.458)
Constant -3.845*** -4.059*** -4.687*** -4.526*** -4.191*** -4.225*** -5.008*** -5.240***
(0.317) (0.283) (0.168) (0.202) (0.159) (0.237) (0.416) (0.446)
Observations 24 23 24 23 24 23 24 23
Adj. R-squared 0.99 0.99 0.99 0.99 0.99 0.99 0.98 0.98
F-Statistic 1334.4*** 894.5*** 4351.8*** 2984.9*** 4136.8*** 1545.2*** 808.1*** 375.3***
Notes: Columns 1 and 2 present the results for the total tax revenue productivity. Columns 2 and 4 presents the
results for the income tax revenue productivity. Columns 5 and 6 present the results for PAYE while columns 7 and
8 presents the results for the VAT revenue productivity. The dependent variable is computed at the annual level. All
regressions are estimated with product and month fixed effects. Robust standard errors are in parenthesis below the
estimated coefficients. *** p<0.01, ** p<0.05, * p<0.1

5.3. Dummy variable regression: measuring tax buoyancy

As discussed in section 2, Lesotho has gone through a number of tax policy reforms. These
reforms have had an impact on the revenue productivity of the country’s tax system. Table 4
presents the results of the buoyancy of the Lesotho tax system.

Table 4: OLS regression results on tax buoyancy in Lesotho (1992-2015)

(1) (2) (3) (4) (5) (6) (7) (8)
VARIABLES

Log (GDP) 1.210*** 1.252*** 1.221*** 1.248*** 1.233*** 1.256*** 1.227*** 1.253***
(0.040) (0.032) (0.035) (0.031) (0.027) (0.020) (0.048) (0.053)
Log (GDP -1) -0.107 -0.168 -0.535 -0.292
(0.716) (0.434) (0.463) (0.538)
Dummy (Tax-rate) 0.008 -0.109** -0.070 -0.132***
(0.090) (0.038) (0.055) (0.011)
Dummy (Tax-bracket) 0.029 0.112 -0.043 0.025
(0.083) (0.073) (0.042) (0.026)
Dummy (Tax-
structure) -0.020 0.020 0.147*** 0.145***
(0.079) (0.062) (0.026) (0.038)
Dummy (Tax-credit) 0.055 0.066 0.074 0.052
(0.048) (0.059) (0.043) (0.041)
Dummy (LRA) 0.047* 0.053*
(0.024) (0.027)
Dummy (VAT2003) 0.010 0.011
(0.033) (0.040)
Dummy (VAT2014) 0.055 0.044
(0.041) (0.049)
Constant -3.756*** -4.134*** -3.848*** -4.084*** -4.584*** -4.738*** -4.968*** -5.193***
(0.368) (0.294) (0.328) (0.297) (0.243) (0.197) (0.456) (0.527)

14
Observations 24 23 24 23 24 23 24 23
Adj. R-squared 0.98 0.99 0.99 0.99 0.98 0.98 0.97 0.97
F-Statistic 413.2*** 596.2*** 1248.1*** 3356.5***
Notes: Columns 1 and 2 present the results for the total tax revenue productivity. Columns 2 and 4 present the results
for the income tax revenue productivity. Columns 5 and 6 present the results for PAYE while columns 7 and 8
present the results for the VAT revenue productivity. The dependent variable is computed at the annual level. All
regressions are estimated with product and month fixed effects. Robust standard errors are in parenthesis below the
estimated coefficients. *** p<0.01, ** p<0.05, * p<0.1

Information in Table 4 shows that in general, the system is buoyant as revealed by the
statistically significant coefficients of GDP which are more than unity. The introduction of the
Lesotho Revenue Authority (LRA) in 2003 seems to have improved the efficiency of the tax
system in general. With the inclusion of other reforms, the lags in policy implementation become
insignificant. Tax credit is also not statistically significant except in the F-Statistics (indicating
its importance in the model). This result confirms the view that tax credit is a policy intended for
equity of the tax system and not revenue productivity. The results also reveal that the transition
from General Sales Tax to Value Added Tax does not seem to have improved revenue
productivity for the economy. Similarly, the expansion of the turnover threshold is not
significant since it was meant to relieve small companies and not for tax revenue productivity.
This confirms the conclusion that policy reforms on VAT have not been revenue productive over
the period of analysis.
In general, when we account for tax reforms, the introduction of LRA appears to have improved
the overall productivity of the Lesotho tax system although marginally. However, the tax reforms
thus far seem to be inclined towards equity and not efficiency objective of the tax system.
Therefore tax revenue in Lesotho grows due to an increase in income and not due to government
effort in tax collection.

Further analysis
We estimated tax elasticity and buoyancy across longer periods with the argument that measures
of tax buoyancy are likely to vary significantly from year to year, which is not very helpful
(Jonathan Haughton, 1998). It is thus more useful to measure buoyancy over a longer period -
perhaps five years at a time.

15
 6. POLICY IMPLICATIONS

This study estimated the revenue productivity of the Lesotho tax system. Based on the results,
there is potential for VAT to raise more revenue for the government. However, future reforms
will have to be targeted towards encouraging investment in the private sector that has more
impact on economic growth. Although the equity focus of the tax policy is commendable, there
is need to shift focus toward revenue productivity such that future reforms are geared towards
encouraging the efficiency in tax collection.

References
Adari, M.M. (1997). Value Added Tax in Kenya. M.A. Research Paper, University of Nairobi.
Artus, K.K. (1974). Tax Revenue Forecasting: A Methodological Study with Application to
Turkey. Studies in Domestic Finance, no. 5 Washington, D.C.: IBRD.
Bayu, T. (2015). Analysis of Tax Buoyancy and Its Determinants in Ethiopia (Cointegration
Approach). Journal of Economics and Sustainable Development, 6(3), 182-194.
Bonga, W.G., N.L. Dhoro-Gwaendepi, and F. Mawire-Van Strien (2015). Tax Elasticity,
Buoyancy and Stability in Zimbabwe. IOSR Journal of Economics and Finance, Vol
6(1), pp. 21-29.
Botlhole T.D. and Agiobenebo T. J., (2006). The Elasticity and Buoyancy of Botswana Tax
System and their Determinants. ICFAI Journal of Financial Economics, Vol. 4(4), pp.
48-62.
Chand, S.K, and B. Wolf (1973). The Elasticity and Bouyancy of the Tax System of Peru,
1960-71: an Empirical Analysis. IMF, unpublished.
Cotton, J.J. (2012). The Buoyancy and Elasticity of Non-oil Tax Revenues in Trinidad and
Tobago (1990-2009), Working paper, WP 06/2012.
Ehdaie, J. (1990). An Econometric Method for Estimating the Tax Elasticity and the Impact on
Revenues of Discretionary Tax Measures. World Bank Working Papers, No. 334.
Howard, M. (1992). Public Finance in Small Open Economies: The Caribbean Experience,
Praega Publishers, CT.
Kargbo, B. I. B. and Egwaikhide, F. O. (2012). Tax elasticity in Sierra Leone: a time series
approach. International Journal of Economics and Financial Issues, Vol. 2(4), pp. 432.
Khan, M.Z. (1973. Responsiveness of Tax Yields to Increase in National Income. Pakistan
Development Review, Vol. XII, no. 4.
Leuthold, J. and T. N’Guessan. (1986). Tax Buoyancy and Elasticity in Developing Economy”
Bureau of Economic and Business Research. Faculty Working Paper No. 1272, Urban
Champain: University of Illinois.
Milambo, M. (2001). Elasticity and Buoyancy of the Zambia Tax System, Unpublished M.A.
Paper, University of Nairobi.
Mukarram, F. (2001). Elasticity and Buoyancy of Major Taxes in Pakistan. Pakistan Economic
and Social Review. Vol. 39 (1), pp. 75-86.
16
Muriithi M.K. and E. D. Moyi, (2003). Tax Reforms and Revenue Mobilization in Kenya, AERC
Research Paper 131, Nairobi: AERC.
Musgrave, R.A. and Musgrave, P.B. (1984). Public Finance in Theory and Practice, 4th edition.
Singapore: McGraw-Hill Book Company.
Mwega, F.M. (1986). The Incidence of Taxes and Transfers in Kenya: A General Equilibrium
Analysis. East African Economic Review, Vol 2,pp 6- 3.
Nambiar, K.V. and Govinda R. M. (1972). Tax Performance of States. Economic and Political
Weekly, pp. 1036-1038.
Okech, T. C and Mburu P. G., (2011). Analysis of Responsiveness of Tax Revenue to Changes
in National Income in Kenya between 1986-2009. International Journal of Business and
Social Science, Vol. 2, No. 21, Special Issue, November
Osoro, N.E. (1992). Revenue Productivity of the Tax System in Tanzania, 1979-1989. Journal of
African Economies, Vol. 1(3), pp. 395-415.
Osoro, N.E. (1993). Revenue Productivity Implications of Tax Reform in Tanzania. African
Economic Research Consortium Research Paper No. 20.
Osoro, N.E. (1995). Tax Reforms in Tanzania: Motivations, Directions and Implications.
African Economic Research Consortium Research Paper No. 38.
Rajkumar, P.V. and S. Chidambaram (1972 ). A Study of the Sensitivity of the Select Taxes in
Tamil Nadu. Southern Economic Review, Vol. 1 (4), pp. 348-356.
Sahota, G.S. (1961). Indian Tax Structure and Economic Development. Asia Publishing House,
Bombay, pp. 7-9
Selialia F.L. (1993) Personal income tax in Lesotho: its impact on the economy. Central Bank of
Lesotho
Singer, N. M. (1968). The Use of Dummy Variable in Establishing the Income Elasticity of State
Income Tax Revenue. National Tax Journal, 21, pp200-204.
Timsina N. (2006) Tax elasticity and buoyancy in Nepal: A Revisit. URL: Vol19_art2.pdf.com

17
Appendix

Table 1: Relative Shares of Main Taxes in Total Tax Revenue in Lesotho

Property Excise Taxes on Other SACU

Year PAYE Taxes GST/VAT Taxes Trade Taxes receipts
1992 11.165 2.814 12.717 4.021 0.000 0.128 53.220
1993 10.040 2.100 11.680 2.591 0.011 0.095 57.179
1994 9.524 0.004 10.398 2.255 0.000 0.209 58.594
1995 9.654 2.672 9.742 2.813 0.002 0.177 50.157
1996 11.678 1.542 7.177 2.208 0.003 0.140 53.566
1997 11.542 0.000 10.541 1.919 0.004 0.088 53.268
1998 12.629 1.636 10.989 1.838 0.003 0.071 50.592
1999 13.477 2.100 11.193 1.386 0.004 0.062 55.010
2000 12.167 0.000 11.219 1.107 0.004 0.077 48.153
2001 13.659 1.788 10.393 1.240 0.003 0.113 49.049
2002 12.857 1.356 10.965 1.688 0.002 0.109 45.306
2003 15.241 1.450 14.096 1.472 0.001 0.212 42.008
2004 13.665 0.000 15.621 1.270 0.144 0.073 44.825
2005 13.509 0.000 14.432 1.701 0.519 0.065 50.006
2006 11.706 1.049 12.835 1.336 0.748 0.101 52.308
2007 9.002 0.841 10.385 1.384 1.231 0.092 60.264
2008 9.758 0.865 10.966 0.978 1.362 0.111 53.739
2009 8.779 0.000 10.751 1.363 0.911 0.069 52.381
2010 12.075 1.057 13.349 1.253 2.084 0.020 36.633
2011 14.905 1.135 13.851 1.766 3.825 0.270 28.028
2012 11.734 1.143 12.813 1.528 0.999 0.004 40.735
2013 11.158 1.265 12.308 0.959 2.175 0.041 44.034
2014 11.758 0.679 14.558 1.402 3.306 0.009 48.136
2015 11.661 0.917 14.334 1.709 1.565 0.009 43.411
2016 11.305 0.924 14.684 1.857 1.661 0.008 42.512
Average 0.118 0.011 0.121 0.017 0.008 0.001 0.485
Note: the table presents the percentage share of individual tax revenue to total government tax revenue between
1992 and 2016

18