1. OPEC+’s “Reasonable Oil Price Level” Notion and the External Breakeven in Saudi Arabia, Russia and Canada: Accounting for Economic Cycles and Pipeline Politics
United States Association for Energy Economics (USAEE) Conference, 2019
Dr. Noha A. Razek, The University of Regina
Dr. Nyakindi Michieka, University of California, Bakersfield
Prof. Emilslon Silva, The University of Alberta

2. Outline Background Why it is important The Objective Literature Review
Contribution to Existing Research
The Model
Methodology
Main Findings
Conclusion

3. Background
Declaration of Cooperation: OPEC+ seeks to create a sustainable stable oil market & establish a reasonable oil price that benefits all industry stakeholders & the global economy.
Yet the question is, what is a reasonable oil price? In this paper, we attempt to answer this question for Saudi Arabia, Russia & Canada.

4. Why it is Important
Why study external breakeven not just fiscal breakeven?
Why study Saudi, Russia, and Canada?
Why assess the impact of geopolitics and pipeline politics?
Why study cyclical movements?

5. Why it is Important – Why assess oil exporters’ external breakeven?
In addition to fiscal strain, oil exporters’ external balances come under pressure & CAs deteriorate when oil prices drop. Although some oil exporters can finance external deficits through their external wealth, others face pressure on their reserves.
The CA for a country with limited export diversification & a prominent oil sector is inextricably tied to the oil balance, creating a systematic relationship between the CA & oil prices.
Managing their foreign assets would enable oil exporters to mitigate abrupt declines in oil production & exports, & make their economies more resilient to oil price shocks.

6. Why it is Important – Why assess oil exporters’ external breakeven
Why it is Important – Why assess oil exporters’ external breakeven? (Cont’d)
Chalk & Hemming (2000) showed the relationship between fiscal & external sustainability is neither 1-to-1 nor entirely independent. Fiscal & external sustainability are not perfect substitutes unless oil producers are fully owned & operated by the government.
Benchmarking would help policymakers determine the magnitude of policy adjustments needed to attain external breakeven (finance their external deficits and obligations to the rest of the world).
(Gnimassoun et al., 2017; Behar & Fouejieu, 2016; Versailles, 2015; Allegret et al., 2014; Arezki & Hasanov, 2013; Tabeke & York, 2011; Blanchard & Milesi-Ferretti 2009).

7. Why it is Important – Why study Saudi, Russia and Canada?
The year 2018 was eventful for the oil market, demonstrating that OPEC matters & so do other non-OPEC producers. Presently, the largest producers in the market are Saudi, Russia, & the U.S., followed by Canada (EIA, 2019).
Saudi & Russia play leading roles in OPEC+ & the global oil market.
Canada is an interesting case because it has close trade links with the U.S. & a more diversified oil exporting economy (Gnimassoun et al., 2017) & is not an OPEC+ member.
Zooming in on oil producing provinces, AB, which is heavily oil dependent, faces the same economic issues as Saudi & Russia.

8. Russia has faced international sanctions and pipeline politics; and
Why it is Important – Why assess the impact of geopolitics and pipeline politics?
Saudi Arabia has endured recent attacks on key infrastructure, including its vital East-West pipeline, oil tankers passing through the strait of Hormuz, and Aramco facilities;
Russia has faced international sanctions and pipeline politics; and
Alberta faces pipeline politics and is landlocked, with more accentuated takeaway obstacles (as reflected in the price differentials depicted in the Figure below).

9. Why it is Important – Why assess the impact of geopolitics and pipeline politics? Cont’d

10. Why it is Important – Why cyclical movements?
The relative inelasticity of supply & demand wrt oil price contributes to the cyclical nature of price movements.
In examining the macroeconomic impact of oil price shocks, Kilian et al. (2009) developed Kilian Index (KI) which captures global business cycle.
The budget deficit is not only due to the loosening of fiscal policy. It could also be due to the economy going through a downturn (state of the business cycle).
If nonlinearity is ignored, policies could be based on seriously flawed inputs.
(Gnimassoun et al. 2017; Büyükşahin et al., 2016; Enders, 2015; Kilian et a., 2009; IMF, n.d.)

11. But what price is reasonable?
The Objective
OPEC+ has set a goal of establishing a stable oil market and achieving a reasonable oil price that benefits all industry stakeholders and the global economy.
But what price is reasonable?
Rather than analyzing the fiscal breakeven or full-cycle breakeven, we examine the effect of the oil price threshold on external balances with the aim of evaluating a country’s ability to finance its external deficits.
We study the cases of Saudi Arabia, Russia and Canada.

12. Bems and de Carvalho Filho (2009)
Literature Review
Bems and de Carvalho Filho (2009)
Morsy (2009), Tabeke & York (2011), Behar & Fouejieu (2016), Allegret et al.(2014) & Akanbi & Sbia (2018) applied panel data analysis to different oil exporters over different time periods.
Their results:
emphasized the close connection between fiscal & external balances & the importance to account for financial development, demographics & nonlinearity;
showed that the fiscal policy has a larger impact than the exchange rate adjustment, which may be ineffective, for oil exporters;
Financial development negatively affects the relationship between the fiscal balance & CA.
They included Saudi Arabia & Russia, but did not provide details on these specific countries in their findings, opting instead to discuss the results for the entire sample.
None of these researchers included Canada in their samples.

13. Literature Review (Cont’d)
Afonso & Rault (2009) & Bluedorn & Leigh (2011) examined the relationship between CAs & budget balances for a group of OECD countries. They discussed findings for the entire sample without providing specifics about Canada & did not study the effect of oil prices.
Gnimassoun et al. (2017) studied the impact on the Canadian CA of oil price movements. Their results highlighted the importance to account for the degree of domestic financial development, the management of foreign exchange reserves & the tendency to spend oil revenues on imports.
Kilian et al. (2009) studied the effects of disentangled oil price shocks on different external balance measures for OPEC & Canada. KI captures the global business cycle. They did not study Russia.

14. Literature Review (Cont’d)
Only Kilian et al. (2009) accounted for economic cycles.
Gnimassoun et al. (2017) & Kilian et al. (2009) did not examine the impact of the fiscal policy.
Bems and de Carvalho Filho (2009), Morsy (2009), Tabeke & York (2011), Behar & Fouejieu (2016), Allegret et al.(2014), Akanbi & Sbia (2018), Gnimassoun et al. (2017) & Kilian et al. (2009) neither studied the impact of pipeline politics nor the 2014 oil price shock.

15. Literature Review (Cont’d)
Neither Setser & Frank (2017) nor Kleinberg et al. (2018) provide statistical modelling-based analysis.
Setser & Frank (2017) analysed what they denote as the ‘external breakeven’. They studied a group of countries – including, Saudi Arabia & Russia but not Canada.
Kleinberg et al. (2018) provided a comprehensive analysis of oil price breakeven points for a tight oil project & discussed fiscal breakeven, but do not examine external balances.

16. We account for cyclical movements using threshold regression analysis.
Contribution
We build on Gnimassoun et al. (2017), and empirically examine the oil price threshold effect on external balances; accounting for fiscal policy, international reserves, propensity to spend oil revenues on imports, exchange rate, pipeline politics, and nonlinearity.
We account for cyclical movements using threshold regression analysis.
We apply time series rather than panel data techniques.
We perform detailed case studies of Saudi Arabia, Russia and Canada.
We study time samples of 1980–2018, 2002Q1–2019Q1, and 2005Q1–2019Q1 for Saudi Arabia, Russia, and Canada, respectively.

17. Methodology – The Model
We model the current account to GDP ratio as follows: 𝐶𝐴 𝑡 𝐺𝐷𝑃 𝑡 =𝑓( 𝑏𝑑𝑔 𝑡 𝐺𝐷𝑃 𝑡 , 𝐵𝑟𝑒𝑛𝑡 𝑡 , 𝑃𝑂𝑑𝑖𝑓𝑓 𝑡 , 𝑅𝐸𝐸𝑅 𝑡 , 𝑖𝑚𝑝 𝑡 , 𝑅𝑆𝑅𝑉 𝑡 ) 𝐶𝐴 𝑡 𝐺𝐷𝑃 𝑡 is the current account to GDP ratio (we also use the change in NFA to GDP ratio ( ∆𝑁𝐹𝐴 𝑡 𝐺𝐷𝑃 𝑡 )); 𝑏𝑑𝑔 t 𝐺𝐷𝑃 𝑡 is the fiscal balance to GDP ratio; Brent 𝑡 is the Brent nominal oil price; POdiff 𝑡 is the oil price differential (Brent-Fateh, Brent-Urals, WTI-WCS); REER 𝑡 the real effective exchange rate; 𝑖𝑚𝑝 𝑡 is the propensity to spend oil revenues on imports; and RSRV 𝑡 is the value of official international reserves.

18. Empirical Methodology – Threshold Regression
Using a threshold value 𝜃, a two-regime threshold regression can be defined without loss of generality as follows:
𝑦 𝑡 =𝐶+ 𝑥 𝑡 𝛽+ 𝑧 𝑡 𝛼 1 + 𝜀 𝑡 if 𝜔 𝑡−𝑑 ≤𝜃
𝑦 𝑡 =𝐶+ 𝑥 𝑡 𝛽+ 𝑧 𝑡 𝛼 2 + 𝜀 𝑡 if 𝜃<𝜔 𝑡−𝑑
𝑦 𝑡 is the dependent variable; 𝑥 𝑡 is a 1𝑥𝑘 vector of regressors ; 𝛽 is a 𝑘𝑥1 vector of regime in-variant parameters; 𝜀 𝑡 is normally distributed error with mean 0 and variance 𝜎 2 ; 𝑧 𝑡 is a vector of explanatory variables with regime-specific coefficients 𝛼 1 and 𝛼 2 ; 𝜔 𝑡 is an observable threshold variable that may be one of the variables in 𝑥 𝑡 or 𝑧 𝑡 (but not necessarily so) and must be predetermined relative to 𝜀 𝑡 ; and 𝑑 is the delay parameter. Regime 1 is the subset of observations where 𝜔 𝑡 < 𝜃, and regime 2 is the subset of observations where 𝜔 𝑡 > 𝜃. Because it may take the regime more than one period to switch, the timing of the regime switch is based on the value 𝜔 𝑡−𝑑 where 𝑑=1, 2,…. 𝐶 is a 𝑘𝑥1 vector of constant terms (Tong & Lim, 1980; Potter, 1999; Bai & Perron, 2003; Enders, 2015; Stata, 2019).

19. Empirical Methodology – Estimation of Threshold Regression (Cont’d)
The dependent variable is the current account to GDP ratio (or an equivalent measure), and
The explanatory variables are the lagged value of the dependent variable and a combination of the explanatory variable of interest.
The variables of interest are reflected in 𝑧 𝑡 and hence are allowed to be variant across regimes.
The dummy variables and seasonal dummy variables are included in 𝑥 𝑡 .
TAR distributed lagged model.
Stigler (2012) argued against excluding the influential threshold variable from the list of regressors. Hence, the oil price variable (i.e., the threshold variable) is included as a regressor.
Referred to as an open loop threshold autoregressive system (Tong & Lim 1980).

20. Robustness Different combinations of variables
Different threshold regression models
Wald Test
Diagnostic tests
Weak and Strong Exogeneity
Capacity utilization
Refinery margin

21. Main Findings
Although Gnimassoun et al. (2017) states the relationship between the CA & oil price could be linear in less diversified economies & nonlinear in more diversified economies, our results show that the relationship is nonlinear in all three countries.
The external breakeven is approximately USD 57–58/barrel for Russia, USD 61– 65/barrel for Saudi Arabia, and USD 74-76/barrel for Canada; i.e., oil prices below these threshold levels compromise these countries’ abilities to finance their external deficits and obligations to the rest of the world.

22. Main Findings (Cont’d)
Our estimates for Saudi Arabia are within the range of the IMF (FRED, n.d.a) external breakeven estimates.
Our estimate for Russia is consistent with Renaissance Capital’s estimated breakeven of USD 56/barrel for the overall fiscal balance (as opposed to the USD 40/barrel for the primary budget balance) (Aris, 2018). It is also close to the 2015 oil exporters’ composite breakeven price of 56 USD/ barrel estimated by Setser and Frank (2017).
These results explain why, in the summer of 2019, Saudi Arabia was not comfortable with USD 60/barrel oil price and was willing to cut production to drive oil prices upward, whereas Russia viewed an average oil price of USD 60–65 USD/barrel as reasonable (Blas, 2018; Nabiullina, 2018; Paraskova, 2019a).

23. Main Findings (Cont’d)
The estimated threshold effect for Canada is close to the 2015 composite measure of the high oil price breakeven of USD 78/barrel estimated by Setser and Frank (2017), and is consistent with Burleton and Abdelrahman’s (2018) argument that the critical factor behind Alberta’s economic recovery has been the increase in crude prices to the range of USD 65–75/barrel.

24. Main Findings (Cont’d)
Geopolitics have detrimental impacts on Saudi Arabia under all oil price regimes.
Russia has managed to weather the impacts of sanctions.
To overcome the possibility that current sanctions could affect Russia’s long-term competitiveness and to diversify the customer base and sources of investment in infrastructure and upstream projects, Russia:
Has been seeking alternate (non-Western) sources of funding and equipment
Has implemented an import substitution policy to incentivize domestic firms to invest in local R&D.
Has been diversifying its exports markets away from the West and cooperating with Asia has become a priority for the Russian government (IEA, 2018).

25. Main Findings (Cont’d)
Canada is negatively impacted by pipeline politics, particularly when the Brent oil price is below USD 47–50/barrel.
An oil price increase has a negative effect on the Canadian current account when oil prices are below USD 47–50/barrel, supporting the “Dutch disease” argument that an oil price increase hurts other Canadian sectors under a low price regime; however,
Under the high price regime (> USD 74–76/barrel), positive impacts of higher oil prices, including increased resource wealth and consumer purchasing are sufficient to overcome the challenges associated with “Dutch disease.”

26. Conclusion
A reasonable oil price is greater than USD 61–65/barrel for Saudi Arabia, USD 57– 58/barrel for Russia and USD 74–76/barrel for Canada.
Russia has been able to weather the effect of Western sanctions and pipeline politics
Canada is negatively impacted by pipeline politics, particularly when the Brent oil price is below USD 47–50/barrel.
In Canada, an increase in oil prices provides benefits that overcome “Dutch disease” under a high oil price regime, but not under a low oil price regime.

27. Figure 4. Saudi Arabia (1980–2018)
Source: SAMA, IMF, World Development Indicators, Oxford Economics, OECD Annual National Accounts, and Thomson Reuters, and The Department of Energy (United Kingdom).

28. Figure 5. Russia (2002Q1–2019Q1)[1]
Source: Central Bank of Russia; Federal State Statistics Service, Russia; Ministry of Finance of the Russian Federation; IMF; Oxford Economics; EIA; and Thomson Reuters.

29. Figure 6. Canada (2005Q1–2019Q1)[1]
Source: Oxford Economics, CANSIM/Statistics Canada, Department of Finance Canada, IMF, Energy Information Agency (EIA), Eikon/Thomson Reuters, and the Government of Alberta.

30. IMF definition: “A current account is unsustainable if it cannot be financed on a lasting basis with market-based capital inflows and it is not consistent with adequate growth, price stability and the country’s ability to service fully its external debt obligations”