Selected Issues

Abstract

Selected Issues

Inflation Dynamics in São Tomé and Príncipe1

The note examines the dynamics and determinants of inflation in São Tomé and Príncipe from 2002 to 2017, a period during which the exchange regime was switched from a flexible one to a fixed peg with the euro. It shows that both supply-side factors, such as international food and oil prices as well as weather conditions, and demand-side factors, such as government spending and tourist arrivals, influenced inflation. The peg has helped stabilize inflation.

A. Introduction

1. The Democratic Republic of São Tomé and Príncipe is a micro-state with the classic characteristics that challenge such countries. Located in the Gulf of Guinea, São Tomé and Príncipe is sub-Saharan Africa’s smallest country by land surface (<1,000km2) and second smallest by population (<200,000). It shares many of the defining characteristics of micro-states, including remoteness, fragility, limited resources, susceptibility to climate and external shocks, and aid dependency. For instance, fluctuating weather patterns (especially rainfall) has a non-negligible impact on agriculture and food prices; in particularly fresh vegetables, such as tomato, are more sensitive to flooding than other crops.

2. The dual-island economy has been largely based on subsistence agriculture. The dense mountainous country is full of fruits and the ocean has plentiful fish. Before independence, Portuguese-owned plantations occupied 90 percent of the cultivated area. After independence, control of these plantations passed to various state-owned agricultural enterprises. The main crop on São Tomé is cocoa, representing about 95 percent of agricultural exports. Other export crops include copra, palm kernels, and coffee. Staple foods include fish, seafood, rice, wheat, beans, maize, and breadfruit. Tropical fruits, such as pineapple, avocado and bananas, are a significant component of the cuisine. The country imports most of its main food items such as rice, wheat, beans, and maize. It has been making inroads into tourism development but needs to upgrade key infrastructure to support the industry. The government is the main formal employer for the population, which has a substantial proportion of young people (>60 percent under 25 years). External assistance drives the formal economy and the country imports most of its consumption and investment goods.

3. The country had steady growth and decelerating inflation in recent years. During 2002-2017, growth averaged 4.7 percent (Table 1). While growth has slowed somewhat from an average of 5.2 percent during 2002-2009 to 4.3 percent during 2010-2017, one key benefit of the euro peg adoption since January 2010 has been the reduced growth volatility. The standard deviation fell from 2.8 (pre-peg) to 0.3 (post-peg). Meanwhile, annual inflation has moderated significantly from an average of 20.8 percent during 2004-2008. Supported by the peg, inflation has fallen sharply from the historic high of 28 percent in 2007 down to 4.6 percent in 2016, the lowest in the past two decades.

Table 1.

São Tomé and Príncipe: Real Gross Domestic Product and Consumer Prices

(Annual Percent Change, 2002-17)

article image
Sources: IMF, World Economic Outlook database; and IMF staff calculations.Notes: For country groups, the methodology was 1) first calculate the average value for each country for the period in question; 2) then calculate either the average or the median of the group of countries in question, using the values obtained in the previous step. Data may not be available for all countries, for the whole time period. Averages for each country are computed using the earliest possible data.

4. However, inflation in São Tomé and Príncipe has been relatively high compared with the euro area and its main trading partners. The annual inflation in São Tomé and Príncipe between 2010 and 2017 was 8 percent, almost 7 times as high as that in the euro area during the same period (1.2 percent). This could be partially attributed to the “Balassa-Samuelson Effect,” that is, countries experiencing faster growth tend to experience real exchange rate appreciation, which would manifest itself in higher inflation under a pegged exchange rate regime. Nevertheless, the real appreciation associated with this relatively large inflation gap could erode the competitiveness of the economy. Indeed, the External Balance Assessment (attached to the staff report) finds the real exchange rate in São Tomé and Príncipe to be moderately overvalued. An examination of the drivers and evolution of the inflation dynamics could help guide polices towards achieving the external balance.

B. Background

5. The consumer price index (CPI) has been strengthened recently. Even though it is compiled using data only from the Agua Grande district—which contains the capital city of the country, it represents well the market prices in the country, because trading in the main island, São Tomé, takes place mostly in Agua Grande. The CPI basket was expanded, rebased and reweighted as of December 2015, with the support of IMF technical assistance. There are 423 products in the index across 12 categories in the following order of importance by relative weight: food and non-alcoholic beverages (hereafter called ‘food’); housing, water, electricity, gas etc.; transportation; clothing and footwear; alcoholic beverages and tobacco; furnishing and household equipment; health; recreation and culture; communication; restaurants and hotels; miscellaneous goods and services; and education.

6. Imported food prices have contributed the lion’s share to inflation from December 2016 to June 2017, after which locally-produced food prices became a more important contributor due to unusual rainfall patterns. For instance, annual inflation spiked unexpectedly to 7.7 percent at end-2017, driven by increased import taxes on alcoholic drinks and meat as well as a temporary shortage of locally produced foods. In particular, more-than-usual rain during the dry seasons damaged vegetable crops and impeded small boat-based fishing.

7. Anecdotal information and early data show that the currency redenomination in January 2018 did not have a significant impact on inflation. The central bank redenominated the local currency (dobra) on January 1, 2018 by removing three zeros. The currency is now valued at 24.50 new dobras per euro, maintaining the peg adopted in 2010. Post-redenomination price data points are not yet sufficient to clearly confirm the impact of redenomination.2

8. The inflation in the country is not driven by indexation in the economy, which is confirmed by a January 2018 CPI mission conducted by the IMF’s Statistics Department (IMF, 2018). The mission did note that the government takes into account the inflation rate when reviewing certain benefits and administered prices.

C. Selected Literature Review

9. A large empirical literature has examined the determinants and dynamic properties of inflation in sub-Saharan Africa. Barnichon and Peiris (2007) study the determinants of inflation by examining the interaction among inflation, output gap, and money gap. Caceres, Poplawski-Ribeiro and Tartari (2011) find that imported commodity prices and governments are two main driving forces of the inflation dynamics in the Central African Economic and Monetary Community (where the latter occurs mainly through controlled prices and the role of capital expenditure in domestic activity).3 Focusing on the period before Malawi’s exchange rate regime switch in 2012, Mangani (2012) finds that cost-push inflation played a more important role than demand-pull inflation. Focusing on the post regime-switch period in Malawi, Wu (2017) examines the role of the pass-through of the exchange rate and policy determinants in driving inflation in Malawi, and finds that after the switch to a floating exchange rate regime in 2012, nonfood prices affects both the headline inflation and the food inflation. Nguyen and others (2015) find that in recent years (since the mid-2000s), the contribution of the supply shocks to inflation has fallen; instead, domestic demand pressures as well as global shocks have played a larger role in driving inflation.

D. Data and Approaches

10. The analyses use quarterly data from 2002 Q4 to 2017 Q4, which cover both the pre-and post-peg periods. This enables us to examine the potential impact of the structural break (i.e., the introduction of the exchange rate peg in January 2010) on inflation.

11. Descriptive analysis is conducted to decompose the headline inflation and to understand the contributions of various components. To focus on the recent pick-up in inflation, the decomposition of overall inflation is conducted for January 2015-December 2017, and the decomposition of food inflation is conducted for December 2016-December 2017 (due to data availability).

12. In addition, time series analysis is conducted to quantify the impacts of the underlying driving factors of inflation, including both supply-side and demand-side factors. This analysis is conducted over the entire sample period based on quarterly data to capture the medium-to-long-term dynamics of inflation, and is done separately for food inflation and non-food inflation to account for the different driving forces of these two. The independent variables are selected based on existing literature, which suggested that inflation is usually affected by characteristics such as the proportion of oil and food imports, vulnerability to weather shocks, economic importance of agriculture, trade openness and policy regime (Wu, 2017). Based on STP data, for food inflation, the supply-side factors include the international food price index and local rainfall4; the demand-side factors include broad money (M2) growth and domestic primary spending (capturing public demand), as well as the growth in the number of tourist arrivals, which generate demand for local goods and services. For non-food inflation, the supply-side factors include international oil price index, given the country’s heavy reliance on oil imports for energy, and the demand-side factors are kept the same as in the food inflation analysis. In both food and non-food analyses, a dummy variable to indicate the exchange regime switch is included, so is the nominal effective exchange rate (NEER) that captures the potential substitution effect.

E. Results from Descriptive Analysis: Headline Inflation Decomposition

13. The food sub-index has been the biggest contributor to overall inflation (Figure 1). An analysis of the contributions to year-on-year inflation by category in each month since January 2015 (when the data decomposition first became available) reveals that food prices are by and large the main contributor. This observation is not surprising given the high relative share (73 percent) of food in the consumption basket. Month-to-month changes are volatile but usually positive, with only 4 instances of negative rates during January 2015 to December 2017. Monthly inflation rarely exceeded 1 percent; however, 4 of the 5 such cases were recorded during 2017.

14. Food price inflation is affected mostly by locally-produced food items, specifically vegetable and fish. Within the food sub-index, the product categories with the largest weights (cereals, vegetables and fish) are also the ones that drive food price inflation (year-on-year basis), particularly the latter two in recent time because of unusual rainfall pattern. Looking at the sources of production, we observe that locally-produced food (mainly fish, fresh fruits and vegetables) has been a key driver to food inflation since mid-2017; previously imported food was the driver of inflation in São Tomé and Príncipe.

F. Results from Time Series Analysis: Drivers of Medium-to-Long-Term Inflation

15. The exchange rate peg has lowered both food inflation and non-food inflation, highlighting the important dampening effect of the peg on inflation. The regression results for food and non-food inflation are shown in Tables 2 and 3, respectively. In each case, the results are presented across three columns, using: (1) the pooled data across the entire sample period; (2) data in the pre-peg period; and (3) data in the post-peg period. The first columns in the two tables show that the peg has lowered inflation significantly: by 7.2 percentage points for food inflation, and by 8.7 percentage points for non-food inflation. This dampening effect is likely to have transmitted both through the discipline imposed by the peg on the central bank’s money issuance, and through the signaling effect that has stabilized economic agents’ expectations of future inflation.

16. Both food inflation and non-food inflation are positively correlated with M2 growth in the post-peg period, suggesting a demand-pulling effect of fiscal spending on inflation. As the third columns of both tables show, M2 growth is highly significant and positively correlated with both food inflation and non-food inflation. Given the dominant role played by the government in the economy, M2 growth is closely associated with fiscal spending as illustrated in Figure 2. Therefore, this result indicates that fiscal policies play a significant role in the inflation dynamics. In this context, a natural question would be why not include fiscal spending directly as an independent variable in the regression. Because of the close linkage between fiscal spending and M2 growth, including both variables would cause multicollinearity. In addition, the short time series of fiscal spending data makes it difficult to identify the true effect of fiscal spending on inflation.5 Furthermore, M2 is preferable as a broader measure than fiscal spending, as it captures the effects of grants that are not channeled through the budget. Therefore, the team decided to use M2 growth among the two as an independent variable.

17. Food inflation is significantly affected by international food prices and weather conditions, highlighting São Tomé and Príncipe’s heavy dependence on food imports and its low agricultural productivity. As shown in Table 2, the international food price growth is positively correlated with food inflation across all three columns. This is consistent with the fact that the lion’s share of the food that São Toméans consume is imported. In addition, in the post-peg period, local rainfall is found to have pushed up food inflation, by disrupting the production in the fishing and other agricultural sectors. With small-scale production and low technology, the agriculture sector is highly vulnerable to weather conditions. It also partially explains the high inflation in the second half of 2017, since the country had an unusually high rainfall during that period.

Table 2.

São Tomé and Príncipe: Food Inflation Regressions (2002 Q4 – 2017 Q4)

article image
Notes: p-values in parentheses. *** p<0.01, ** p<0.05, * p<0.1All results are corrected for autocorrelation (up to 4 lags) and heteroskedasticity using the Newey-West estimator. Source: IMF staff calculations.

18. Non-food inflation is positively correlated with growth in tourist arrivals, pointing to bottlenecks in the development of local infrastructure to support the industry. As shown in Table 3, tourist arrivals’ growth is positively correlated with non-food inflation (note that the data on tourist arrivals are only available since 2012 Q1, in the post-peg period). This result is intuitive as a larger number of tourists tends to increase the demand for local services and raise non-food inflation. However, it can also be interpreted as a sign of the country’s supply-side bottlenecks of local services. This interpretation is consistent with the hotel room shortages during peak times as noted by some hotel managers.

Table 3.

São Tomé and Príncipe: Non-Food Inflation Regressions (2002 Q4 – 2017 Q4)

article image
Notes: p-values in parentheses. *** p<0.01, ** p<0.05, * p<0.1All results are corrected for autocorrelation (up to 4 lags) and heteroskedasticity using the Newey-West estimator. Source: IMF staff calculations.

19. Graphical analyses confirm the results in the time series analysis. Figure 3 plots some key variables with food and non-food inflation. Figure 3a shows that food inflation broadly follows the trend of M2 growth in the post-peg period. For example, both series experience a decrease around 2011 Q3, and both experience an increase around 2012 Q3. Figure 3b shows that food inflation follows the trend of growth in international food prices relatively closely. Finally, Figure 3c shows that non-food inflation (blue solid line) follows both the trend of M2 growth (orange solid line) and that of growth in tourist arrivals (red dotted line). Note that each of these figures only captures the univariate relationship between one factor (e.g., M2 growth) and inflation, therefore a perfect alignment between these series and inflation are not expected.

G. Conclusion

20. Our empirical results demonstrate explicit linkages between macroeconomic policies and inflation. Both supply and demand factors affect the inflation dynamics in São Tomé and Príncipe, and government policy could play a role in influencing inflation and guiding inflation expectations. Specifically, to entrench price stabilization, the following polices could be pursued:

  • Support the pegged exchange rate system with coherent macroeconomic policies. As the analysis demonstrates that the peg has been beneficial in stabilizing inflation, it is important to support the peg. To do so, the reserves buffer needs to be strengthened. This, in turn, could be achieved through a package of policies, including more closely monitoring the implementation of externally-financed projects to avoid delays in grant disbursements, as well as the demand-side and supply-side policies stated below.

  • Continue fiscal consolidation to manage domestic demand. The analysis demonstrates that inflation is significantly impacted by M2 growth, which, in turn, is closely associated with the growth of public expenditure. Continued fiscal consolidation would help ease the local demand pressure and support the building of an international reserves buffer.

  • Increase local supply, enhance productivity, and improve infrastructure:

    • Utilize more local products and decrease the dependence on food imports. This could be achieved by adjusting the consumption habits (e.g., a greater reliance on indigenous staples, such as breadfruit and green banana, instead of imported rice), and by enhancing the processing of local primary agricultural products.

    • Improve agricultural productivity. This could be done by building greenhouses, promoting sustainable agriculture, and making agriculture more robust to weather shocks. The ongoing joint effort in this area by the authorities with other partners, including China and the United Nations, is a positive step.

    • Build infrastructure and increase the capacity of local services to support tourism demand. In this context, the implementation of the tourism development strategy that the authorities launched in January 2018 with the support of the World Bank should help eliminate the supply-side bottlenecks on tourism over time.

Figure 1.
Figure 1.
Figure 1.

São Tomé and Príncipe: Contributions to Overall and Food Inflation

Citation: IMF Staff Country Reports 2018, 322; 10.5089/9781484385005.002.A002

Sources: São Tomé and Príncipe authorities and IMF staff calculations.
Figure 2.
Figure 2.

São Tomé and Príncipe: M2 Growth and Domestic Primary Spending Growth, 2008 Q1 - 2017 Q4

(Percent)

Citation: IMF Staff Country Reports 2018, 322; 10.5089/9781484385005.002.A002

Sources: São Tomé and Príncipe authorities and IMF staff calculations.
Figure 3.
Figure 3.
Figure 3.

São Tomé and Príncipe: Inflation and Various Factors

Citation: IMF Staff Country Reports 2018, 322; 10.5089/9781484385005.002.A002

Sources: São Tomé and Príncipe authorities and IMF staff calculations.

References

  • Barnichon, R. and S. J. Peiris, 2007, “Sources of Inflation in Sub-Saharan Africa,” IMF Working Paper 07/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Caceres, C., M. Poplawski-Ribeiro, and D. Tartari, 2011, “Inflation Dynamics in the CEMAC Region,” IMF Working Paper 11/232 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2018, São Tomé and Príncipe – Report on Consumer Price Index Mission (January 8-19) (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mangani, R., 2012, “The Effects of Monetary Policy on Prices in Malawi,” AERC Research Paper 252 (Nairobi: Africa Economic Research Consortium).

    • Search Google Scholar
    • Export Citation
  • Nguyen, A. D. M., J. Dridi, F. D. Unsal, and O. H. Williams, 2015, “On the Drivers of Inflation in Sub-Saharan Africa,” IMF Working Paper No. 15/189 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Wu, D. F., 2017, “Understanding Inflation in Malawi: A Quantitative Investigation,” IMF Working Paper No. 17/48 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

1

Prepared by Marlon Francisco, Jehann Jack and Yunhui Zhao.

2

Consumer prices inched up consecutively during 2018 Q1 so that cumulative inflation, at 1.1 percent, was noticeably lower (120 basis points) at end-March 2018 compared to the rate reported for the corresponding period last year. The dobra redenomination in January 2018 may not have led to any price gouging effects.

3

As explained later, our note has similar findings.

4

Specifically, it is the percentage deviation of each quarter’s rainfall from the long-run quarterly mean since 1901.

5

Quarterly fiscal spending data are available only since 2008 Q1.

Democratic Republic of São Tomé and Príncipe: Selected Issues
Author: International Monetary Fund. African Dept.