Kiribati
Selected Issues and Statistical Appendix
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar Close
Publication Date:
30 Jun 2009
eISBN:
9781451821901
Language:
English
Keywords:
ISCR; CR; Kiribati; cost; fee; RERF balance; Kiribati's authorities; RERF assets; RERF drawdown; asset mix; RERF level; Stocks; Government debt management; Bonds; Pacific Islands
Download PDF (1.4 MB)

This Selected Issues paper focuses on recent developments with Kiribati’s Revenue Equalization Reserve Fund (RERF). The paper also examines fiscal aspects of climate change, and considers options for improving fishing license fees, which remain an important source of revenue. It also analyzes recent developments and the outlook for remittances to Kiribati, which is another important source of external revenue and brings important economic benefits, such as reducing poverty and stabilizing national income.

Abstract

This Selected Issues paper focuses on recent developments with Kiribati’s Revenue Equalization Reserve Fund (RERF). The paper also examines fiscal aspects of climate change, and considers options for improving fishing license fees, which remain an important source of revenue. It also analyzes recent developments and the outlook for remittances to Kiribati, which is another important source of external revenue and brings important economic benefits, such as reducing poverty and stabilizing national income.

I. The Revenue Equalization Reserve Fund: Recent Developments and Outlook1

A. Background

1. The RERF is a key source of income for Kiribati. The fund was established in 1956 and capitalized using phosphate mining royalties. Under a conservative policy, the fund’s assets grew steadily until 1979 when phosphate deposits were exhausted. In subsequent years, this conservative approach continued, but since 2001 there has been an increase in fiscal deficits and drawdowns from the RERF. The fund has taken a relatively risk-averse investment strategy, and the current share of bonds in the portfolio is around 70 percent.

Table I.1.

RERF Assets, Selected Years

article image
Source: Authorities and IMF Staff Estimates.

B. Recent Trends and Projections

2. More recently, the RERF has been adversely affected by the global financial crisis. In 2008, the capital loss in A$ terms was over 10 percent of the fund’s outstanding value;2 and drawdowns (for financing the budget deficit) mean the per capita RERF balance dropped by around 20 percent to less than A$4,200 (in 1996 A$). This is less than 60 percent of the peak in 2000 (of over A$7,000).

uA01fig01

RERF per capita balance (In 1996 A$)

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Sources: Kiribati’s authorities and staff estimates.

3. Going forward, the outlook for the fund has worsened. If past trends persist, even under a relatively optimistic assumption of 6 percent nominal returns, the fund is projected to be depleted by 2030—five years earlier than the projection during the 2007 Article IV consultation (using similar assumptions).3

4. To maintain the per capita RERF balance (in 1996 A$), a substantial fiscal effort is required. An informal target for the RERF used by the authorities in the past has been to maintain the real per capita value of the RERF at or above the level in 1996 (A$4,700). Under the baseline parameter values, targets for the real per capita level of the RERF near the current level or a simple fiscal sustainability rule suggest fiscal deficits need to be reduced to around 6 percent of GDP, implying an adjustment of around 6 percent of GDP from the expected deficit this year. (Baseline parameter values: nominal returns of 6.0 percent; real GDP growth of 1.1 percent; and an inflation rate of 2.8 percent).

Table I.2.

Sustainable Fiscal Deficits and RERF Drawdowns

article image
Source: IMF Staff Estimates.

5. These projections are subject to significant uncertainty, particularly if a higher share of equities in the portfolio is chosen. To quantify the uncertainty, stochastic simulations are performed in a simple model with four variables: real returns, real GDP growth, the inflation rate, and RERF drawdowns (in percent of GDP). The model allows for correlations of the shocks to the four variables.4 Results are shown in the fan charts (the center, middle and outer areas cover 50 percent, 70 percent, and 90 percent confidence intervals, respectively). Mean nominal returns from equities are 7.0 percent, mean nominal returns from bonds are 5.0 percent, and real GDP growth and the inflation rate are the same as those in the nonstochastic simulation above.5

6. The simulation results confirm the importance of a conservative fiscal stance for sustainability of the RERF. Regardless of growth rates and returns, the faster pace of fiscal drawdowns (as in recent years) is likely to be unsustainable. At the same time, the higher growth scenario highlights the benefit of an increase in tax revenues.

Table I.3.

Summary of the Stochastic Simulation Results

article image

In this scenario, an increase in tax revenue (partly) offsets fiscal drawdowns. Assumed drawdowns of 15 percent and 7.5 percent are those before subtracting this offsetting effect.

On the other hand, assuming a 10 percent mean nominal return, the per capita balance is projected to be maintained with a probability of around 75 percent.

Figure I.1.
Figure I.1.

Per Capita RERF Balance (In 1996 A$): Baseline Scenario

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Source: Kiribati’s authorities and staff estimates.
Figure I.2.
Figure I.2.

Per Capita RERF Balance (In 1996 A$): Higher Growth Scenario (3.1 Percent Real GDP Growth)

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Source: Kiribati’s authorities and staff estimates.
Figure I.3.
Figure I.3.

Per Capita RERF Balance (In 1996 A$): Lower Return Scenario (3.5 Percent Nominal Returns)

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Source: Kiribati’s authorities and staff estimates.
Figure I.4.
Figure I.4.

Per Capita RERF Balance (In 1996 A$): Higher Return Scenario (8.0 Percent Nominal Returns)

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Source: Kiribati’s authorities and staff estimates.

C. Implications of the Large Capital Loss and Reform Directions

7. The asset mix of the fund is currently around 30 percent equity (from around 50 percent equity before 2005).6 The asset mix is less risky than other funds in Pacific island countries and is not a primary reason for the decline of the fund. Indeed, Kiribati’s RERF still averaged a 4.5 percent return during 2004–08. Moreover, simulation results using historical data on equity and bond returns indicate that, if we take 1990 as the starting point, the realized level of the RERF is much higher than in the 100 percent bonds portfolio; and nearly the same level in 2008 as a 100 percent equities portfolio.

Table I.4.

Share of Equity in Reserve Fund

(In percent)

article image
Sources: Authorities and IMF staff calculations.

8. In addition to a reduction in fiscal drawdowns, there is a need for a well-defined target or rule for the RERF combined with a medium-term budgeting framework. Successful experiences of natural resource funds (for example in Norway) indicate that formulating a medium-term plan is effective in allowing expenditure smoothing. If reforms can reduce fiscal uncertainty, a more return oriented portfolio mix might be appropriate.

uA01fig02

RERF Levels (In millions of A$)

Citation: IMF Staff Country Reports 2009, 196; 10.5089/9781451821901.002.A001

Sources: Kiribati’s authorities and staff calculations.

Some Lessons on Drawdowns and Longer-term Fiscal Planning

Successful cases:

Tuvalu

  • The Tuvalu Fund’s independent advisory committee consists of qualified economists, financial experts and others.

  • The committee has not only provided advice to trustees’ decision making, but also has helped to incorporate the fund’s management into long-term fiscal planning, taking into account the impact of fiscal drawdowns on the fund (ADB, 2005).

Norway

  • Norway formulated a well-defined numerical rule, which limits drawdowns for the non-oil deficit to the realized real returns in the fund. This helps to cap the drawdowns from the fund to a sustainable level.

  • Further, the operation of the rule is supported by fiscal projections covering a 50-year period. The forecasts help reduce political pressures for higher spending (and larger drawdowns), (IMF, 2005).

Unsuccessful case:

Mexico

  • While Mexico had numerical rules on how excess revenues to the fund should be used, these have been relaxed over time, and there is no medium-term fiscal framework. As a result, the authorities failed to build up temporary windfalls from oil receipts (IMF, 2005).

9. Efforts to improve governance, particularly through better disclosure would also be welcome. Experiences elsewhere indicate that weak governance will likely to lead to poor performance of a reserve fund.

Governance, Disclosure and Fund Performance

Successful case:

Norway

  • The public is well informed about how the money is invested and what the returns have been. Detailed information is also readily available to the public.

  • This high degree of transparency has helped to build a consensus in decision making at both policy and operational levels (IMF, 2005).

Unsuccessful cases:

Tonga

  • The operation of the fund has been outside of the fiscal system, and has not been subject to the same accountability as the national budget.

  • In the late-1990s, the fund incurred a US$26 million (20 percent of GDP) capital loss due to risky investments (following the investment strategy of an advisor who was subsequently sued by the authorities). The structurally weak fiscal setup is believed to have increased the potential for the misuse of resources (ADB, 2005).

Nauru

  • The governance on the fund had been very weak; the fund was used as collateral for borrowing against (which was violation of the rules), and poor investments were made. As a result, the fund was (nearly) depleted by 2005, (ADB, 2005).

References

  • Graham, Benjamin, 2005, “Trust Funds in the Pacific,” (Manila: Asian Development Bank).

  • Purfield, Cartriona, 2005, “Managing Revenue Volatility in a Small Island Economy: The Case of Kiribati,” IMF Working Paper http://www.imf.org/external/pubs/ft/wp/2005/wp05154.pdf (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
1

Prepared by Kiichi Tokuoka.

2

The RERF capital loss in 2008 was 30 percent in U.S. dollar terms, but the large depreciation of the A$ by end-2008 contained the RERF capital loss in A$ terms.

3

With a conservative assumption of 3.5 percent nominal returns, the fund is projected to be exhausted in 2026. On the other hand if a higher return is assumed, the timing of the depletion will be postponed. For example, with an 8 percent nominal return (roughly the historical average over 1990–2008), the fund is projected to be exhausted in 2036.

4

As a first step, a Vector Autoregression is run using the four variables, and the variance-covariance matrix for the error terms, denoted Ω, is estimated. Then, simulations were performed using stochastic error terms with covariance structure Ω.

5

This set of parameter values produces the 6 percent nominal return assumed in the non-stochastic simulation above, if the portfolio is 50 percent equities and 50 percent bonds.

6

The share of equities declined to around 30 percent in 2005, because Nikko Asset Management, which manages half the fund, reduced its share of equities to 15 percent (at the maximum) from 50 percent.

Same Series

Other IMF Content

Other Publishers

  • Collapse
  • Expand
Cover IMF Staff Country Reports
Kiribati: Selected Issues and Statistical Appendix
Author:
International Monetary Fund
Volume/Issue:
Volume 2009: Issue 196
Publisher:
International Monetary Fund
ISBN:
9781451821901
ISSN:
1934-7685
Pages:
49
DOI:
https://doi.org/10.5089/9781451821901.002
  • View in gallery

    RERF per capita balance (In 1996 A$)

  • Figure I.1.
    View in gallery
    Figure I.1.

    Per Capita RERF Balance (In 1996 A$): Baseline Scenario

  • Figure I.2.
    View in gallery
    Figure I.2.

    Per Capita RERF Balance (In 1996 A$): Higher Growth Scenario (3.1 Percent Real GDP Growth)

  • Figure I.3.
    View in gallery
    Figure I.3.

    Per Capita RERF Balance (In 1996 A$): Lower Return Scenario (3.5 Percent Nominal Returns)

  • Figure I.4.
    View in gallery
    Figure I.4.

    Per Capita RERF Balance (In 1996 A$): Higher Return Scenario (8.0 Percent Nominal Returns)

  • View in gallery

    RERF Levels (In millions of A$)

RERF per capita balance (In 1996 A$)
Figure I.1.

Per Capita RERF Balance (In 1996 A$): Baseline Scenario
Figure I.2.

Per Capita RERF Balance (In 1996 A$): Higher Growth Scenario (3.1 Percent Real GDP Growth)
Figure I.3.

Per Capita RERF Balance (In 1996 A$): Lower Return Scenario (3.5 Percent Nominal Returns)
Figure I.4.

Per Capita RERF Balance (In 1996 A$): Higher Return Scenario (8.0 Percent Nominal Returns)

RERF Levels (In millions of A$)