Determinants of Aggregate Saving

An Overview

Lemi Taye

A term paper submitted in completion

of an assignment for the course
Advanced Macroeconomics I

Addis Ababa University

College of Business and Economics
Department of Economics
December, 2017
1 Introduction
Saving, a sacrifice of current consumption, provides for the accumulation of capital which, in
turn, produces additional output that can potentially be used for consumption in the future.
The process is thus inherently intertemporal. Its presumed operation makes the saving
behavior of citizens and their governments central to the development of poor countries
(Gersovitz, 1988).
Growth in output of an economy depends partly on accumulation of human and physical
capital, which in turn depends partly on available savings. Saving is a major source to
finance investment which in turn brings economic growth. As a result neoclassical growth
models put saving (and investment) at the core of the growth process (Perkins et al., 2013).
In fact a high saving rate positively affects the real per capita economic growth rate (Adema,
2015; Loayza et al., 2000b). Thus it would be fair to say that economic growth is the result
of abstention from current consumption (Ray, 1998).
Saving not only allows for growth in income and increase in consumption, but also for
the smoothing of consumption in the face of volatile and unpredictable income, and thus
helps to ensure the living standards of poor people whose lives are difficult and uncertain
(Gersovitz, 1988; Deaton, 1989). Studies suggest that savings and poverty have a significant
and negative relationship implying that those who save are more likely to move out of poverty
while those who do not, remain in poverty (Ting, 2013). As a result promoting personal
saving is viewed as a strategy to combat poverty.
Thus the desirability of saving is unquestionable both from a microeconomic and a
macroeconomic perspective. It follows that identifying factors that determine saving is im-
portant particularly for policy purposes. This short paper reviews both theoretical and
empirical literature on the determinants of aggregate saving (from a macroeconomic point of
view). It then extends the discussion by providing descriptive statistics both at the interna-
tional level and focusing on Ethiopia. It concludes by highlighting some policy implications.

2 Literature Review

2.1 Theoretical Literature

The major theories that try to explain saving include Classical theory, Keynesian/Absolute
income hypothesis, Relative income hypothesis, the Life-cycle hypothesis, and the Perma-
nent income hypothesis. The more recent neoclassical theories, i.e., the life-cycle hypothesis
and the permanent income hypothesis, focus on microeconomic foundations to describe con-
sumption and saving.
According to the Classicals, people that receive income do not spend their entire income
on the purchase of consumption goods; they save a part of their current earnings for future
spending. This implies that a part of national income leaks out of the aggregate expenditure
stream in the form of saving (Johnson, 2015). To the classical economists, saving is interest-

1
elastic, and an increase in the rate of interest will have stronger effects of increasing the
marginal propensity to save than discouraging investment (Johnson, 2015; Tiriongo, 2005).
In this theory, the level of investment is constrained by the volume of savings available.
This theory assumes full employment in the economy and therefore the level of income is
exogenous. It also assumes full employment in the economy and therefore the level of income
is exogenous (Tiriongo, 2005).
According to the Keynesian savings framework, savings is a stable and increasing function
of the level of income (Ting, 2013; Tiriongo, 2005). Keynes envisaged the marginal propensity
to save to be governed by what he called objective and subjective factors, and included such
factors as the rate of interest, social and institutional changes, and subjective motives for
savings (Tiriongo, 2005). Keynes (1936) postulated that the average propensity to consume
falls as income rises. This suggests that the average propensity to save, also known as
the saving rate, which is the ratio of saving to disposable income, should rise with income
(Mishkin, 2014). However, this was later challenged by an empirical anomaly which came to
be known as the Kuznets paradox (Parker, 2011; Mankiw, 2015). Keynes (1936) also argued
that saving is not so responsive to interest (Johnson, 2015).
The relative-income hypothesis, described by James Duesenberry (1949), was an attempt
to address the Kuznets paradox. He argued that a households consumption would depend not
just on its own current level of income, but on its income relative to those in the subgroup
of the population with which it identifies itself. The household will attempt to align its
consumption expenditures with those of other members of its group. Thus, households with
lower income within the group will consume a larger share of their income to keep up, while
households with high incomes relative to the group will save more and consume less (Parker,
2011).
The life cycle hypothesis, mainly due to Modigliani and Ando (1957, 1963), posits that
consumption and saving reflect an individual’s stage in the life cycle, which is generally
proxied by age. Since retirement, for most people, is the most substantial and enduring
income fluctuation, this model emphasizes saving for retirement as a primary motivation for
deferred consumption (Beverly and Sherraden, 1999; Mishkin, 2014). The prediction of this
model is that there should be a relation between aggregate saving and the rates of population
growth and income growth. If saving is accumulated during the working years to finance
retirement, as the hypothesis suggests, then population growth provides more savers than
dissevers, and positive aggregate saving (Deaton, 1989).
Friedman (1957) proposed the permanent income hypothesis which originates from the
basic intuition that individuals would wish to smooth consumption and not let it fluctuate
with short run fluctuations in income. This suggests that an individual’s consumption in
a given period is determined not by income in that period, but by permanent or lifetime
income (Meghir, 2004; Romer, 2011). According to this hypothesis observed differences in
household saving and consumption are believed to reflect, in part, differences in the relative
shares of transitory and permanent income. In other words, when an individual experiences
an increase in income that she perceives as temporary, she is expected to save, rather than
consume, this money (Beverly and Sherraden, 1999). Another implication of this hypothesis
for saving is that people save because they rationally expect their labor income to decline

2
and thus saving can be a predictor of declines in labor income (Campbell, 1986).

2.2 Empirical Literature

There are numerous studies conducted on the macroeconomic determinants of saving both
at the international level and considering the case of specific countries. A few of these are
reviewed below.
Loayza et al. (2000a) used a large macroeconomic panel data set consisting of 69 coun-
tries 20 industrial and 49 developing, spanning the years 1966-95 to establish the stylized
facts concerning the effects on the private saving rate of its key policy and non-policy de-
terminants. They estimated a variety of empirical equations for the private saving rate for
a worldwide sample of countries, as well as separately for industrial and developing country
subsamples. They employed panel data procedures, paying particular attention to the issues
of simultaneity and country heterogeneity and used internal instruments to correct for these
problems. Their main empirical findings were: private saving rates rise with the level and
growth rate of real per capita income; dependency ratios have a negative effect on private
saving rates supporting the predictions of the life-cycle hypothesis; inflation has a positive
impact on private saving, holding other factors constant, which supports the precautionary
motive for saving; and fiscal policy is a moderately effective tool to raise national saving.
Masson et al. (1998) investigated the determinants of private saving using a panel data
set covering both industrial and developing countries. For the industrial countries, the panel
data set contains 21 countries over 1971-93 whereas 40 developing countries over 1982-93 were
considered. They regressed saving rates for industrial and developing countries on several
potential explanatory variables that could be collected on a reasonably comparable basis
across all countries. They found that demographics and growth are important determinants
of private saving rates, and that interest rates and terms of trade have positive, but less
robust, effects. Increases in per capita gross domestic product was found to increase saving
at low income levels (relative to the United States) but decrease it at higher ones. They
also found that there is a partial offset on private saving of changes in public saving and (for
developing countries) in foreign saving.
Balassa (1989) studied the effects of interest rates on saving in developing countries
by estimating a time-series model for individual countries and cross-sectional model for a
number of countries. He found that both estimates point to the positive effects of interest
rates on savings. He estimated the interest elasticity of saving for developing countries and
found it to be in the range of 0.1 0.2 percent, i.e., a one percent increase in interest rates
would raise savings by 0.1 0.2 percent, which was relatively lower when compared with
the corresponding figure for the United States. He also found evidence of capital flight in
response to higher interest rates abroad.
Tiriongo (2005), using a time series data, estimated a hybrid model to study the deter-
minants of aggregate domestic private savings in Kenya. The model was based on variables
identified from the life cycle hypothesis, the permanent income hypothesis and the simple
Keynesian framework. The study covered the reform period 1980 to 2003 during which

3
a substantial fluctuation in aggregate savings to GDP ratio was observed. Economic and
demographic factors, and measures of financial sector development were considered in the
study. The results showed that aggregate savings is significantly determined by the current
account deficit, the ratio of M2 money1 to gross domestic product, real per capita income
growth, deposit rate and the old age dependency ratio.
Nwachukwu and Odigie (2011) employed an error correction model to examine the deter-
minants of private saving in Nigeria for the period between 1970 and 2007. The framework
of their analysis involves the estimation of a saving rate function derived from the life cycle
hypothesis while recognizing the structural characteristics of a developing economy. The
result shows that the saving rate rises with both the growth rate of disposable income and
the real interest rate on bank deposits. They also found that public saving does not seem to
crowd out private saving with the implication that government policies aimed at improving
the fiscal balance have the potential of bringing about a substantial increase in the national
saving rate in Nigeria.

1
The M2 monetary aggregate includes currency and demand deposits (together called M1), plus savings
deposits and several less liquid asset types, such as money market deposit accounts and money market
mutual fund shares with check-writing features.

4
3 Data and Discussion
In the world we live in today, significant differences in saving rates can be found between
countries. One can observe differences even in broad classifications of countries according to
their income status. Figure 1 shows the gross saving rates of low income, middle income and
high income countries. Although the data for low income countries is not complete (data is
missing for the years before 1990), it can be seen that their saving rate is fairly low, while
middle and high income countries tend to save more. It also shows that the gap in saving
rates between middle and high income countries has widened since the early 2000s.

Figure 1: Gross Domestic Saving Rates in high, middle and low income countries

So this begs the question: what caused such huge gap in saving rates among countries?
One factor that has been proposed is the rate of per capita income growth. Figure 2 illustrates
the relationship between saving and growth of real GDP per capita for a sample of 84
countries.2 It implies a positive correlation between growth and saving. However, correlation
does not imply causation, meaning we cannot infer whether high saving drives growth or
high income growth leads to high rates of saving.
In the neoclassical growth models (Solow (1956) and Swan (1956)) saving is an important
determinant of growth. That is countries with high rates of saving (and investment) will have
high incomes and vice versa. But recent literature has focused on the simultaneity between
growth and saving, that is, the level of growth may itself affect the saving rate. The primary
2
See Adema (2015), Section 3.2.

5
studies to support the proposition that the rate of growth of per capita income affects the
saving rate were Left (1969) and Modigliani (1970) (cited in Gersovitz, 1988). Ram (1982)
provides the strongest evidence that per capita income has a significant effect on the saving
rate. A relatively recent study by Carroll and Weill (1994) indicates that economic growth
causes domestic saving, but not vice versa.

Source: World Bank Indicators Database cited in Adema (2015).

Figure 2: The relationship between saving and growth

3.1 Gross Domestic Saving in Ethiopia

Accumulation of a sizable capital is one of the prerequisites for structural transformation
for a poor country like Ethiopia. The growth and transformation (GTP I and II) targeted
a higher rate of savings as part of the efforts of domestic mobilization to meet the financial
requirements of large infrastructural projects and investments in the manufacturing sector.
Figure 3 shows that the rate of saving rose steadily during the period 2011/12 2015/16.
Owing to the low level of income and traditional mode of consumption in the rural areas,
the rate of gross domestic savings was low until 2009/10. The rate of gross domestic saving
(gross domestic savings in proportion to GDP) reached 22.4 percent in contrast to the 9.7
percent in 2009/10 (EEA, 2017).

6
 Figure 3: Rate of Gross Domestic Savings and Growth of GDP in Ethiopia

In parallel with the increase in domestic savings, the rate of gross national savings also
reached 32.4 percent in 2015/16. The difference between the two rates was mostly covered
by net current transfer from the rest of the world. In GTP II, resource mobilization from the
domestic private savings and government revenue have been identified as a core component
of financing the development goal. In the last five years, savings have been mobilized mainly
through the government housing scheme and the Grand Renaissance Dam Bond (EEA, 2017).
Other than these forced saving mechanisms policy instruments such as the interest rate were
found to have an insignificant impact on aggregate domestic saving (Getnet, 2017; Haile,
2013).

7
4 Conclusions and Policy Implications
Saving is an important instrument to enhance economic growth by providing sufficient funds
for investment. The importance of saving for growth has been emphasized by neoclassical and
new growth theories. Given the crucial role saving plays to bring about growth in income,
many theories have been proposed to explain what determines saving. The important ones
include the absolute income hypothesis of Keynes (1936), Modigliani’s (1950s) life cycle
hypothesis, and Friedman’s (1957) permanent income hypothesis.
Empirical evidences on the determinants saving in developing countries have identified
many variables that affect aggregate domestic saving. Many studies suggest that growth in
income has a positive and significant effect on saving. In addition, demographic factors such
as the dependency ratio, developments in the financial sector, fiscal policy and budget deficit
were found to affect saving. However, the results for inflation and interest rates are mixed.
Interest rates do not seem to have a significant impact on saving in developing countries.
Several policy implications can be drawn from these analyses. As growth in income is
the primary determinant of saving, efforts has to be made to raise the level of income in
a sustainable manner. Moreover, the budget deficit has to be kept at fairly manageable
levels since it adversely affects growth of saving. There is also a role for prudent fiscal and
monetary policy to encourage private saving. Finally there is a need to urgently develop
the financial sector of the country by further expanding bank branches and services and by
creating a very competitive environment in the financial sector.

8
 References
Balassa, B. (1989). The effect of interest rate on savings in developing countries. World
Bank Working Paper No. 56.

Beverly, S. G., and Sherraden, M. (1999). Institutional determinants of saving: Implica-

tions for low-income households and public policy. The Journal of Socio-Economics, 28(4),
457-473.

Campbell, J. Y. (1986). Does saving anticipate declining labor income? An alternative

test of the permanent income hypothesis. Econometrica, 55(6), 1249-1273.

Carroll, C. D., and Weil, D. N. (1994). Saving and growth: a reinterpretation. In Carnegie-
Rochester conference series on public policy (Vol. 40, pp. 133-192). North-Holland.

Deaton, A. (1989). Saving in developing countries: Theory and review. The World Bank
Economic Review, 3(suppl 1), 61-96.

EEA (2017). Report on the Ethiopian economy: Challenges of sustaining Ethiopias for-
eign exchange earnings from exports and remittances. Addis Ababa, Ethiopian Economics
Association.

Friedman, M (1957). A Theory of the Consumption Function. Princeton University Press.

Gersovitz, M. (1988). Saving and development. Handbook of development economics, 1,

381-424.

Getnet G. (2017). Trends and Determinants of Gross Domestic Saving in Ethiopia. Journal
of Economics and Sustainable Development, 8(5), 72-81.

Haile A. (2013). Determinants of domestic saving in Ethiopia: An autoregressive distributed

lag (ARDL) bounds testing approach. Journal of Economics and International Finance, 5(6),
248-257.

Johnson, A. O. (2015). An Investigation of the Determinants of Savings and Investment

in Nigeria. Issues in Economics and Business, 1(2), 1.

Loayza, N., Schmidt-Hebbel, K., and Servn, L. (2000a). What drives private saving across
the world? The Review of Economics and Statistics, 82(2), 165-181.

Loayza, N., Schmidt-Hebbel, K., and Servn, L. (2000b). Saving in developing countries:
an overview. The World Bank Economic Review, 14(3), 393-414.

Mankiw, G. (2015). Macroeconomics, 9th Edition. Worth Publishers.

9
Masson, P. R., Bayoumi, T., and Samiei, H. (1998). International evidence on the de-
terminants of private saving. The World Bank Economic Review, 12(3), 483-501.

Meghir, C. (2004). A retrospective on Friedman’s theory of permanent income. The Eco-

nomic Journal, 114(496).

Mishkin, F. S. (2014). Macroeconomics: Policy and Practice, 2nd Edition (Pearson Se-
ries in Economics). Prentice Hall.

Nwachukwu, T. E., and Odigie, P. (2011). What drives private saving in Nigeria? African
Economic Research Consortium, Nairobi.

Parker, Jeffrey (2011). Macroeconomic Theory. Reed College (Course book).

Perkins, D. H., Radelet, S., Lindauer, D. L., and Steven, A. Block. (2013). Economics
of Development. W. W. Norton & Co.

Ram, R. (1982). Dependency rates and aggregate savings: a new international cross-section
study. The American Economic Review, 72(3), 537-544.

Ray, D. (1998). Development economics. Princeton University Press.

Romer, D. (2011). Advanced Macroeconomics, 4th Edition. McGrew Hill Higher Educa-
tion.

Solow, R. M. (1956). A contribution to the theory of economic growth. The quarterly

journal of economics, 70(1), 65-94.

Swan, T. W. (1956). Economic growth and capital accumulation. Economic record, 32(2),
334-361.

Ting, L. (2013). Savings and Poverty: Is there really a link? (First working draft for
submission to the ESSA conference). University of Johannesburg.

Tiriongo, S. K. A. (2005). Determinants of aggregate domestic private savings in Kenya,

1980-2003 (Doctoral dissertation).

10