Uruguay: Selected Issues
Author:
International Monetary Fund. Western Hemisphere Dept.
Search for other papers by International Monetary Fund. Western Hemisphere Dept. in
Current site
Google Scholar
PubMedClose
Publication Date:
25 Jan 2022
eISBN:
9798400200335
ISBN:
9798400200335
Language:
English
Keywords:
inflation outcome; inflation level; shift inflation; inflation reduction; deferential vis-à-vis trade partner; Central bank transparency; Fiscal stance; Central bank autonomy; Global
Download PDF (858.7 KB)

This paper studies the impact of structural factors on inflation level and stickiness across Latin American countries. It finds that greater labor market flexibility—reflected in wage setting and decentralized bargaining schemes—leads to lower inflation. Similarly, more disciplined fiscal policies (especially linked to the implementation of fiscal rules) and, to a lesser extent, central bank transparency and independence also help reduce inflation. The results also indicate that implementing reforms along some of these dimensions in Uruguay could have meaningful impact in terms of permanently lowering inflation.

Abstract

This paper studies the impact of structural factors on inflation level and stickiness across Latin American countries. It finds that greater labor market flexibility—reflected in wage setting and decentralized bargaining schemes—leads to lower inflation. Similarly, more disciplined fiscal policies (especially linked to the implementation of fiscal rules) and, to a lesser extent, central bank transparency and independence also help reduce inflation. The results also indicate that implementing reforms along some of these dimensions in Uruguay could have meaningful impact in terms of permanently lowering inflation.

Structural Determinants of Inflation1

This paper studies the impact of structural factors on inflation level and stickiness across Latin American countries. It finds that greater labor market flexibility—reflected in wage setting and decentralized bargaining schemes—leads to lower inflation. Similarly, more disciplined fiscal policies (especially linked to the implementation of fiscal rules) and, to a lesser extent, central bank transparency and independence also help reduce inflation. The results also indicate that implementing reforms along some of these dimensions in Uruguay could have meaningful impact in terms of permanently lowering inflation.

A. Introduction

1. Uruguay’s inflation rate has been one of the highest in the region and persistent throughout business cycles. Inflation around the world and in the region has come down significantly in the past two decades, but Uruguay has been lagging the common trend.

uA002fig01

Inflation: LAC and Uruguay

(in percent)

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

2. High inflation has hampered Uruguay’s financial deepening and eroded competitiveness. Persistently high inflation has been a root cause of high dollarization—as agents have resorted to US dollar assets to hedge against inflation. This, in turn, has resulted in an underdeveloped financial system, with low credit to the private sector (as firms and households are reluctant to borrow in foreign currency). The high inflation deferential vis-à-vis trade partners has also contributed to eroding the country’s external competitiveness.

3. The high and sticky inflation may have structural causes. Conventional business cycle models see inflation as a function of cyclical factors and monetary conditions, such as output gap, unemployment rate, and money supply. Although these factors help describing the cyclical fluctuations of inflation, they do not paint the whole picture about the large and persistent differences in inflation across countries. Slower-moving structural factors such as wage setting mechanism, the degree of fiscal responsibility or the institutional setting of the monetary authority may play a role.

uA002fig02

Dollarization

(In percent)

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

uA002fig03

Credit to private sector

(percent of GDP, 2020)

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

4. Understanding the structural causes of inflation is important for de-dollarization, promoting credit growth, and designing pro-growth policies for Uruguay. The Uruguayan authorities have expressed a clear intention to lower inflation, with efforts so far mostly focused on central bank monetary policy actions, i.e., interest rate setting and monetary policy communication. The structural causes of high inflation, however, may go beyond the mandate of the central bank. More clarity on the non-monetary contributors to inflation can help inform the policy agenda in this regard. This paper examines the role of labor market policy, central banking institutions, and fiscal responsibility as structural drives of inflation in Uruguay and peer countries.

B. How Structural Factors Can Affect Inflation

5. Labor market rigidities can lead to a higher equilibrium level of inflation. Cukierman (1992) proposed that, when wage stickiness is high, the effect from unanticipated monetary expansion for boosting output and employment is relatively large. This gives the authorities a greater incentive to inflate the economy. Over time, private sector agents recognize this policy motive and adjust their price-setting behavior accordingly. Nevertheless, with higher wage rigidity, the incentive to inflate is higher. In equilibrium, the inflation rate is higher where wage stickiness is high, even though the goal of higher employment may not be achieved. Thus, a hypothesis of the paper is that a more flexible labor market, where wages are competitively determined, contributes to a lower (long-run) inflation rate.

6. Backward-looking wage indexation can also lead to higher and stickier inflation. When past inflation becomes a fixed component in wage determination, the wage increase can then feed into future price increases and perpetuate the previous inflation level, other things equal. In addition, backward-looking indexation can make disinflation initiatives costly. Specifically, when nominal wages are dependent on past inflation while current inflation comes down, the real wages increase, which tend to reduce labor demand and output, leading to short-term contractionary effects (Cottarelli et al, 1998). Thus, facing pervasive backward wage indexation, the authorities may need doubly strong political will to implement disinflation initiatives. Extensive indexation may also be a manifestation of labor market rigidities, as the market does not adjust according to micro-level information, but rather, relying on macro indicators which contain less information of firm or sector level performances. Moreover, since greater indexation may reduce the social cost of inflation, it may weaken the economy’s aversion to inflation in general (Fischer & Summers, 1989).

7. The prevalence of centralized wage bargaining may increase both equilibrium inflation level and its stickiness by reducing wage flexibility and inducing backward indexation. A centralized wage bargaining system is inherently more rigid than a decentralized system, as the former has a harder time assimilating locally-originated information on productivity growth and productivity differentials across firms and industries. The higher cost in obtaining new information inputs in a centralized bargaining system may encourage the use of backward-looking wage indexation, since the information on past inflation is easy to obtain compared to other more granular data. In addition, if the legal and institutional environment in a centralized bargaining arrangement increases the bargaining position of large unions, it may lead to higher wages and lower employment (McHugh, 2002) and, in turn, increases the authorities’ incentive to inflate. We therefore hypothesize that a more centralized bargaining process is associated with higher and stickier inflation.

8. The quality of central bank institutions can also affect inflation. Central bank independence and transparency may be more conducive to low inflation. By separating the central bank from a wider agenda of promoting growth and, instead, committing expressly to price stability, these institutions can counteract the inflationary bias of a more discretionary monetary policy. Thus, having a more transparent and independent central bank may be associated with lower inflation.

9. Responsible fiscal policies can support a lower inflation level. Countries with a large negative fiscal imbalance may have a stronger incentive to use inflation as an intangible tax to help alleviate fiscal problems (Cottarelli et al, 1998). In addition, deficit spending may bid up labor costs, which, combined with wage rigidity, may perpetuate a higher inflation level.2 Thus, one can expect a positive relationship between the government’s fiscal deficit and the level of inflation, as well as between public debt and inflation. On the flip side, institutions that support fiscal discipline, such as fiscal rules, may be associated with a lower inflation level. Responsible fiscal management may be especially important for countries dependent on commodity exports. Case studies (e.g. Holden, 2013) show that conservative fiscal policy plays a large role in regulating the upward pressure on domestic demand and price/wage inflation created by a booming export sector in commodities or natural resources. We therefore hypothesize that lower public debt and government deficit, as well as the presence of fiscal rules, should be associated with lower inflation.

C. Data and Stylized Facts

10. The analysis focuses on 18 Latin American countries. The countries included are Argentina3, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, and Venezuela.4 The time period covered is 1995–2018, though some of the structural variables are available for shorter periods, as explained below.

11. Data on labor market institutions, central bank institutions, and fiscal responsibility come from various sources.5

  • Labor market dynamics. The paper uses an index on “flexibility of wage determination” by Global Competitiveness Report (GCR). The index is based on the cross-country survey question: “Wages in your country are set by a centralized bargaining process (=1) or up to each individual company (=7)”. In other words, the higher the index, the more frequently that wages are determined by individual employers. However, the current batch of GCR data is only available starting 2007. The paper also uses the indicator for “centralized collective bargaining”, from Fraser Institute’s Economic Freedom Index. This indicator is essentially based on the same survey from GCR, but it harmonizes data from earlier years, when the survey question was phrased somewhat differently, to get a longer time series. This indicator runs from 2000. Fraser Institute also publishes a composite index on the “flexibility of labor market regulations”. This is an aggregation of several sub-indices that measure how employer-friendly labor market regulations are, including i) minimum wage regulation, ii) hiring and firing regulation, iii) centralized collective bargaining, iv) hours regulation, and v) mandated cost of worker dismissal.

  • Central bank institutions. Two indices are used: The first is an index on central bank independence by Garriga (2016), although it is only updated to 2012. The second is an index on central bank transparency by Dincer and Eichengreen (2014), which runs from 1998 to 2014. The latest data for these two indices are then extrapolated to 2018 using the last actual observation available.6

  • Fiscal responsibility. The paper considers the impact of fiscal deficit, public debt-to-GDP, and the presence of fiscal rules on inflation. The data on primary structural balance from WEO is used. We use the structural balance measure in order to mitigate the influence of the business cycle, which affects both inflation and the fiscal balance. The fiscal rules data is taken from the IMF’s Fiscal Rules and Fiscal Councils Database, which flags four types of rules: budget balance rules (BBR), debt rules (DR), expenditure rules (ER), and revenue rules (RR). We derive a composite fiscal rule indicator by adding the four together. In other words, the value equals 0 if a country does not have any fiscal rule and equals 4 if all four types of rules are present. Although the database is carefully constructed, it has limited country coverage and only include 10 of our sample countries. In addition, the database is only updated to 2014. Thus, the last data observations are extrapolated to 2018.

uA002fig04

Labor Market Dynamics

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

uA002fig05

Central Bank Institutions

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

uA002fig06

Fiscal Responsibility

Citation: IMF Staff Country Reports 2022, 017; 10.5089/9798400200335.002.A002

Table 1.

Uruguay: Summary Statistics of Main Variables

article image

12. Simple cross-sectional correlations show a significant negative relationship between labor market flexibility and inflation. As indicated in the charts—which plot partial regression of a country’s average inflation on its average score of labor market indicators, without control variables—economies with more centralized collective bargaining schemes tend to have higher inflation. In other words, more flexibility of wage determination is associated with lower inflation. The overall flexibility of labor market regulations is also associated with lower inflation, although the negative relation in the chart seems largely driven by one outlier (Venezuela).

13. Some measures related to the central bank’s institutional setting seems associated with inflation, although data limitations are important. Central bank transparency appears negatively correlated with inflation, although there is no clear link between central bank independence and inflation. As mentioned earlier, however, the data for the latter index is only updated to 2012. Thus, results should be interpreted with caution. In addition, the within-country time variation in the indicator is minimal, which makes identification more difficult in a regression setting, as discussed in the next section.

14. Fiscal discipline seems to be associated with lower inflation. Countries running a higher structural primary balance, on average, appear to have lower inflation on average. The presence of fiscal rules is also associated with lower inflation although the relationship, as expressed through simple correlation, does not appear significant. As for the average public debt level, the correlation plot does not show any clear relationship with inflation.

D. Empirical Method

15. We formally explore the impact of structural factors on inflation in a multi-variate setting, along the lines of Cottarelli et al (1998), and Kandil et al (2009). Specifically, inflation is modeled as a function of past inflation, as well as cyclical and structural factors, as follows:

πi,t=αi+βπi,t−1+γXi,t+δsi,t+ϵi,t(1)

where πi,t, is current inflation of country i, si,t is the structural factor under investigation, and Xi,t is a set of control variables mostly related to business cycle and monetary conditions. The latter include 1) output gap and primary fiscal balance to capture the business cycle’s impact on inflation; 2) monetary base as percent of GDP, to represent the monetary stance, 3) import-to-GDP ratio, to capture the degree of openness of the economy, and 4) world inflation, to capture the impact of external inflation trend. We expect the coefficients for 1) and 2) to be positive. For 3), the hypothesis is that a more open economy with a higher import ratio would tend to have a lower inflation (Romer, 1993). The rationale is that the more open the economy, the smaller the real benefits of boosting output from surprise monetary expansion, as the welfare loss from the real terms of trade loss dominates. Thus, a lower equilibrium rate of inflation may be observed. For 4), the coefficient is expected to be positive, as part of the domestic inflation may be imported. Including this control also helps capture the impact of the common long-term shift in the global inflation environment, which has had a distinctive downward trend during the past decades.

16. The model is estimated using OLS and system GMM. The fixed effect estimation for dynamic panels that include lagged dependent variables as regressors lead to biased estimates for the lagged variable (Nickell, 1981), especially in the scenario where the time dimension is not significantly larger than the dimension of the panel. To avoid this issue, we employ the system GMM estimator by Blundell and Bond (1998), which is an extension of Arellano and Bond (1991).

E. Results

Structural Factors and Inflation

17. Labor market institutions appear to be important determinants of the level of inflation (Table 2). The estimates for wage flexibility and the absence of centralized collective bargaining are negative and significant for both OLS and GMM estimators, although the coefficients are larger for the GMM estimator. The centralized collective bargaining indicator covers a longer time series, so we consider it a preferred measure to the first one. Taking the numerical estimates at face value, the results indicate that a 1-unit increase in the wage flexibility score (i.e. 1-unit reduction in centralized collective bargaining score) leads to around 20 percent reduction in inflation level.7 The variable for “flexibility of labor market regulations” is not significant in either estimator.

Table 2.

Uruguay: Labor Market Flexibility and Inflation1/

article image

For all tables, standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

18. Regarding the control variables, the world inflation has the expected sign, while the others are mostly not significant. In most regressions, the output gap has a positive sign, as expected. But contrary to our prior, the import ratio variable has mostly positive signs, albeit not significant in most of the specifications. Higher primary balance8 is shown to be mostly associated with higher inflation, though the estimates are not significant. This variable could be picking up the business cycle variation (i.e., when the economy is booming, fiscal performance tends to improve while inflation also tends to be higher) that is not effectively captured by the output gap variable. None of the estimated coefficient for base money is significant.

19. Central bank institutions also play a role in determining the level of inflation (Table 3). The indicators on central bank transparency and independence are both negative and significant in the OLS regression—i.e., higher scores in central bank transparency and independence lead to lower inflation. However, while the coefficient for central bank independence remains negative, only the transparency variable is significant in the system GMM setting. As mentioned earlier, since these are long-term institutional variables, they do not change much from year to year. The lack of within-country variation may have made identification harder in the GMM setting since it relies on first differences of variables. In addition, the time coverage for both variables are shorter than our full sample. The later years’ observations are filled by extrapolating earlier years’ data. This probably also reduced identification. In terms of magnitude of impact, the GMM estimate indicates that a 1-unit increase in the central bank transparency score is associated with 10 percent reduction in inflation.

Table 3.

Uruguay: Central Bank Institutions and Inflation

article image

20. Fiscal institutions also affect inflation, though some results are not statistically significant. The results indicate that, consistent with the hypothesis, higher structural deficits and public debt levels lead to higher inflation, while the introduction of fiscal rules reduces inflation. However, only the public debt variable is significant in both OLS and GMM settings. The fiscal rule variable is significant in the GMM specification, while primary structural balance is not significant in either setting. One explanation for the lack of significance for primary balance is that a higher deficit in any single year may not be a true indication of a lack of fiscal discipline, thus its relationship with equilibrium inflation is weak. In contract, the debt level is the accumulated result of fiscal management over an extended period of time, thus a more informative indicator of fiscal discipline. As for fiscal rule, aside from the fact that fiscal rules impose material constraints on fiscal management, the presence of fiscal rules itself may be a signal that the authorities are taking fiscal responsibility seriously. In terms of magnitude, according to the GMM estimation results, a 10 percentage point increase in the debt-to-GDP ratio is associated with 10 percent increase in inflation, other things equal, while implementing one type of fiscal rules reduces inflation by around 18 percent.

Table 4.

Uruguay: Fiscal Responsibility and Inflation

article image

21. Strengthening some of Uruguay’s institutional features could contribute to significantly reduce the level of inflation. Table 5 presents the estimated magnitudes of inflation reduction if Uruguay’s scores in the select structural variables were at the sample average levels for LAC countries. While these estimates should be interpreted with caution, including because they were explored in the econometric estimations one at a time, they give a sense of the importance of tackling structural features as a way to reduce inflation. Based on that and on the estimated impacts of various structural variables on inflation, we can calculate what the inflation rate would be if Uruguay’s levels for these structural variables were at our sample average of LAC countries. The table shows that changes in labor market dynamics would likely yield the biggest impact on inflation for Uruguay, followed by improving central bank transparency and implementing fiscal rules.9

Table 5.

Uruguay: Estimated Potential Reduction in Inflation 1/

article image

From the 2020 inflation level of 9.8 percent.

Structural Factors and Inflation Stickiness

22. One way that structural factors can affect inflation is by increasing its persistence. For example, as discussed earlier, backward wage indexation to past inflation can make it harder for future inflation to adjust downwards, as wages and prices feed into each other. It also increases the output/employment cost of any policies attempting to reduce inflation, as real wages will increase in the short term. Thus, if the labor market institutions encourage wide-spread wage indexation, inflation may become stickier. Lack of central bank transparency and independence can also lead to stickier inflation, as the central bank may lack the effective tool and/or resolution to change the status quo and shift inflation to a different equilibrium.

23. To explore the effect of structural factors on inflation stickiness, we modify the baseline specification to include an interaction term, between lagged inflation and the underlining structural factor:

πi,t=αi+βπi,t−1+γXi,t+δsi,t+θπi,t−1⋅si,t+ϵi,t(2)

Typically, inflation exhibits stickiness to a certain extent, manifested as the positive coefficient β for the lagged inflation. If a structural factor increases (decreases) such stickiness, θ, the coefficient of the interaction term between past inflation and the structural factor, should be positive (negative).

24. Tables 6–8 present estimation results of Equation 2 for the structural factors of concern. Table 6 shows that the interaction term is negative and significant for wage flexibility and centralized collective bargaining in both OLS and GMM specifications. The interaction term for labor market regulations is also negative, but it is insignificant in either specification. Overall, this lends support to the hypothesis that one of the ways flexible wages can lead to lower inflation is by making inflation less sticky.

Table 6.

Uruguay: Labor Market Flexibility and Inflation Stickiness

article image
Table 7.

Uruguay: Central Bank Institutions and Inflation Stickiness

article image
Table 8.

Uruguay: Fiscal Responsibility and Inflation Stickiness

article image

25. Table 7 reports the estimates for central bank institutions with the interaction term added. Again, due to the lack of within-country variation and shorter time coverage for these variables, clear identification is harder to achieve. The interaction term is negative for central bank transparency, but the coefficient is insignificant. For the central bank independence score, the interaction term is negative and significant in the GMM estimate but is positive and insignificant in OLS.

26. Table 8 reports similar estimates for fiscal responsibility variables. None of the interaction terms are significant for primary balance, debt level, and fiscal rules. Not only that, it appears adding the interaction term also makes the structural factor itself lose identification, in the case of fiscal rules and debt level. Overall, the results suggest that even though fiscal discipline tends to lead to lower inflation as shown in the results in Table 4, the mechanism likely does not involve reducing inflation stickiness.

F. Conclusion

27. Various structural factors can affect inflation outcomes. This paper investigates whether a country’s inflation level is affected by structural factors related to the flexibility in wage setting and labor market regulations, central bank institutions, and fiscal discipline. Our focus is on Uruguay and other LATAM countries, which share many similar external shocks and business cycle patterns. The paper finds evidence that more flexible wage determination, greater central bank transparency and independence, as well as greater fiscal discipline are all associated with lower inflation, to various extents.

28. Tackling certain structural features, including some beyond the control of the central bank, would support Uruguay’s disinflation plan. In particular, enhancing the flexibility of wage setting mechanisms (to allow for greater decentralization), strengthening central bank transparency and independence, and increasing fiscal disciple (including through an effective implementation of the fiscal rule) could have important bearing on achieving the BCU’s disinflation plan.

References

  • Christoffel, Kai, and Tobias Linzert. “The role of real wage rigidity and labor market frictions for inflation persistence.” Journal of Money, Credit and Banking 42, no. 7 (2010): 1435–1446.

    • Search Google Scholar
    • Export Citation

  • Cottarelli, Carlo, Mark Griffiths, and Reza Moghadam. The Nonmonetary Determinants of Inflation: A Panel Study. IMF Working Paper 98/23, 1998.

    • Search Google Scholar
    • Export Citation

  • Cukierman, Alex. “Central Bank Strategy, Credibility, and Independence: Theory and Evidence.” Journal des Économistes et des Études Humaines 3, no. 4 (1992): 581–590.

    • Search Google Scholar
    • Export Citation

  • Dassatti, Cecilia, and Gerardo Licandro. “Measuring monetary policy transparency in Uruguay.” DT BCU (2020).

  • Dincer, N. Nergiz, and Barry Eichengreen. “Central bank transparency and independence: updates and new measures.” (2013).

  • Fischer, Stanley, and Lawrence H. Summers. “Should governments learn to live with inflation?.” The American Economic Review 79, no. 2 (1989): 382–387.

    • Search Google Scholar
    • Export Citation

  • Garriga, Ana Carolina. “Central bank independence in the world: A new data set.” International Interactions 42, no. 5 (2016): 849–868.

    • Search Google Scholar
    • Export Citation

  • Gwartney, James, Robert Lawson, and Seth Norton. Economic freedom of the world: 2008 annual report. The Fraser Institute, 2019.

  • Holden, Steinar. “Avoiding the resource curse the case Norway.” Energy Policy 63 (2013): 870–876.

  • Kandil, Magda E., Hanan Morsy, and Enrique A. Gelbard. “Determinants of Inflation in GCC.” IMF Working Papers 2009, no. 082 (2009).

  • McHugh, James. “Wage centralization, union bargaining, and macroeconomic performance.” IMF Working Paper (2002).

  • Mello, Miguel, and Jorge Ponce. “Fiscal policy and inflation expectations”. Banco Central del Uruguay Working Papers (2021).

  • Schaechter, Andrea, Tidiane Kinda, Nina Budina, and Anke Weber. Fiscal Rules in Response to the Crisis: Toward the “Next-Generation” Rules: A New Dataset. International Monetary Fund, 2012.

    • Search Google Scholar
    • Export Citation
1

By Natasha Che.

2

Mello and Ponce (2021) found a positive correlation between budge deficit and inflation expectations.

3

Inflation data for Argentina is taken from the World Economic Outlook as reported by the country authorities.

4

Venezuela is not included in the sample used for the regression results presented in later sections of the paper, given that the country is an outliner in many of the structural valuables as well as in inflation. The results do not change qualitatively with Venezuela included.

5

The LAC average is calculated over 2010–18, including all sample countries except Venezuela. For Uruguay, 2018 data is used.

6

Dassatti and Licandro (2020) compared Uruguay’s scoring in different cross-country measures of central bank transparency, and noted that Uruguay’s scores have improved in more recent periods.

7

Note that this is percentage reduction from current inflation level, not reduction in the percentage point of inflation rate.

8

This is the raw primary balance as percent of GDP, not the structural balance.

9

We did not take into account the fiscal rule that was recently implemented in Uruguay, since the rule is new. The calculation of the fiscal rule gap in Table 5 assumes no fiscal rule for Uruguay.

  • Collapse
  • Expand
Cover IMF Staff Country Reports
Uruguay: Selected Issues
Author:
International Monetary Fund. Western Hemisphere Dept.
Volume/Issue:
Volume 2022: Issue 017
Publisher:
International Monetary Fund
ISBN:
9798400200335
ISSN:
1934-7685
Pages:
51
DOI:
https://doi.org/10.5089/9798400200335.002

Inflation: LAC and Uruguay

(in percent)

Dollarization

(In percent)

Credit to private sector

(percent of GDP, 2020)

Labor Market Dynamics

Central Bank Institutions

Fiscal Responsibility