Home | About LCPS | Contact | Careers

Featured Analysis

Amer Bisat, Senior Portofolio Manager and head of Sovereign and Emerging Markets (Alpha) at BlackRock.

May 2020
The Government Economic Plan: A Complicated Maze of Policy Trade-offs

A Government Economic Plan was approved by the cabinet on 30 April, and it is refreshingly rigorous and analytical. Missing so far is a document that analyzes the situation holistically and examines the trade-offs necessary when searching for a way out of the financial crisis.[i] The society and policymakers needed the document’s honest arithmetic of the size and nature of the financial hole and the sacrifice required to fill it. The plan also demystified a number of important taboos: It internalized the notion that the country cannot do it alone and that the IMF route is inevitable; it elegantly justified why the debt moratorium—and restructuring—was necessary; it shed light on the central bank’s insolvency; and nicely explained why a foreign exchange market devaluation was necessary.
However, implicit in the plan’s analysis is a maze of policy choices and trade-offs which this essay will attempt to decode. A careful reading cements two key conclusions. First, the size of the problem is stunningly large. Second, there are no easy solutions. Painless measures are nowhere near enough. And policies that can make a difference (deep fiscal austerity; large devaluations; deposit haircuts; or usage of state assets) are either unbearably painful or socio-politically impossible to implement in today’s Lebanon.
The brief concludes with some preliminary thoughts on gradualism versus radicalism.
The Growth Backdrop
Central to any economic blueprint is how it foresees a country’s future growth path. After all, sustainable prosperity is the ultimate objective of development programs. The plan’s one-to-three years growth forecast is hard to quarrel with. The wrenching economic adjustment implicit in its policy proposals will virtually guarantee a deep recession over the next few years.[ii] Once the recession is over, the document foresees a timid “potential growth” of 3% per year well into the medium term.
What do those forecasts mean at an intuitive level? The 2020-2022 economic contraction, combined with a massive real exchange rate depreciation, will shrink GDP—when measured in USD—by half. More alarmingly, using the paper’s potential growth assumption, Lebanon will not regain its 2018 GDP level in USD until 2043.[iii]
Said differently, the plan, as envisaged, is effectively asking the society to halve its standard of living now while giving scant hope for eventual recovery. It does not take brilliance to predict that any program that produces this kind of growth contour is politically incredibly onerous. The upfront pain is too large and the subsequent recovery too timid. Something must give.
Public Finances
The plan echoes the mantra that the Lebanese budget must shoulder a significant part of the adjustment. Specifically, the assumed reduction in public debt is predicated on the assumption that the government needs to sharply raise its primary surplus to 3.8% of GDP by 2024 and maintain it at that level year in/year out. I would submit that Lebanon doesn’t have the fiscal space needed to generate this kind of surplus.
Start with the spending side. Over the past 15 years, if one excluded interest payments on debt, the Lebanese government spent an average of 21% of GDP on general public services. As it is, that level of non-interest spending is low by international standards and is arguably one reason why Lebanon ranks very low in surveys of “government effectiveness”.[iv] But then, on top of that, the plan prescribes draconian austerity well into the future, foreseeing spending dropping to 17.1% of GDP by 2024, including the assumed CEDRE funded spending. Translated into actual US dollars, this means the government that was spending $11 billion in 2019 will spend $5.7 billion in 2024. That kind of contraction in government services is economically and socially very difficult to imagine as being sustainable.
To be clear, the above isn’t an argument about the quality of current spending, which is incontrovertibly inefficient, wasteful, and in desperate need of rationalization. And the plan rigorously and nicely catalogues the areas where spending can be dramatically reduced. However, it largely ignores areas where more spending is needed. The list of legitimate and currently absent fiscal spending is lengthy and almost certainly larger than the savings that can be obtained from Electricité du Liban and the civil service wage bill. For example, limiting capital spending solely to CEDRE funding, as the document suggests, is too restrictive. Even under the unreasonable assumption that none of the assumed CEDRE money ($0.6 billion annually) goes to social spending, that amount is well short of the government’s capital—including infrastructure—needs.
The revenue side is equally challenging. Over the next two years, the fiscus will be losing the two sources of revenues it has come to heavily rely on: Financial sector taxation and import levies. Partly as a result, the plan sees the revenue-to-GDP ratio at a timid 18.7% by 2024. To raise the ratio to where it “should” be, [v] new taxes would almost certainly be necessary. But the experience with the “WhatsApp tax” which sparked the October protests clearly illustrates how difficult raising taxes will likely prove to be.
There is a final point to be made on the plan’s fiscal section: The document ignores potential future oil and gas revenues and says little about usage of state assets. While I applaud the paper’s reluctance to incorporate uncertain future revenues and to wade into the toxic topic of privatization, the reality is, no serious analysis of Lebanon’s medium-term fiscal sustainability should ignore these potentially constraint-easing resources.
The Financial Sector, BdL’s Balance Sheet and Deposits
The plan estimates the size of the hole in the consolidated financial sector—both banks and Banque du Liban (BdL)—at LBP 186 trillion. Using a LBP 3,500/$ exchange rate, this translates into a $53 billion. This figure is nothing short of staggering.[vi]
It is worthwhile decomposing that top line number. Start with the banks alone: The document implies that banks’ losses—emanating from the restructuring of their holdings of government debt and from impaired private sector debt—stand at LBP 65 trillion.[vii] Notably, banks’ capital is LBP 31 trillion. Put differently, wiping bank owners’ equity (as, reasonably, recommended by the plan) would go a long way to cover banks’ losses. The residual—“only” LBP 34 trillion or $10 billion—while large, is not insurmountable.
Now, moving on to BdL: The numbers are, alas, much larger. The hole in the central bank’s balance sheet—emanating from the restructuring of its holdings of public sector debt and, by far more importantly, from accumulated losses on “financial engineering” operations—stands at LBP 177 trillion. The document, reasonably, makes a number of “allowances” (including accounting for BdL’s equity), but even then, the residual BdL hole is LBP 121 trillion.
The government’s plan asserts that the system’s hole must be filled immediately. That effectively means that banks’ claims on BdL will have to be written off which, in turn, means that banks won’t be able to honor their own depositors. To state the obvious, this is an enormous ask. The plan is effectively asking banks’ depositors to shoulder a 170% of GDP hit on day one.
The plan recognizes the enormity of the task and flirts with ideas to assuage the pain including “clawing back” ill-gotten funds and forcing bank owners to re-inject capital. It also frames depositor participation as voluntary—rather than a haircut—and also recommends a Public Asset Management Company whose profits would accrue to BdL. These are all great ideas, but the reality is their magnitudes are nowhere close to enough. It is inevitable depositors will have to take a large hit.
The question that needs to be raised is whether the draconian recommendation of dealing with BdL losses on day one is necessary. If the alternative is a huge haircut on deposits, shouldn’t we consider the alternative of delaying BdL’s recapitalization? While it is certainly optimal to clean the central bank’s balance sheet, I am not convinced it must be done immediately. Unlike commercial banks, central banks don’t experience “deposit runs”. Moreover, a central bank can create liquidity—in a way that commercial banks cannot—and, as such, it can deal with LBP claims when they come due.
Let me be clear though: Delaying the recapitalization of BdL comes with enormous costs. An insolvent central bank is one that cannot back the local currency, which means that confidence in the LBP will not return anytime soon. Delaying recapitalization will also preclude banks from being able to access their own deposits at BdL—especially those in USD—which, in turn, means the depositors’ freeze currently in place will persist, probably for years. Finally, the only way BdL can create LBP, to honor banks’ LBP withdrawals, without letting the foreign exchange market “go” is by delaying the easing of capital controls. These are all serious and very costly downsides. But that cost must be weighed against the consequences of the massive and instantaneous depositors’ haircut.
Debt Sustainability
The document’s debt section posits that the anticipated debt restructuring operation brings the debt-to-GDP ratio from 176% in 2019 to 103% of GDP in 2020. That reduction comes as a result of an assumed large haircut on the debt’s face, but, against that, takes into account the assumed new foreign borrowing from the IMF.[viii] The plan then foresees broad stability in that ratio for five years before projecting debt monotonically declining thereafter.
This path needs to be deconstructed since it implicitly makes important policy choices.
To begin with, I—and many others—have argued that, for a country like Lebanon, the sustainable debt ratio should be closer to 60% of GDP. Even by the plan’s primary fiscal surplus forecasts, the country will not achieve that debt ratio until about 2032. Put differently, the plan forecasts the country will live with an unsustainable debt stock for an additional 12 years. For that whole period, interest rates will remain high, inflows sparse, and liquidity tight. In other words, growth will suffer.
There is another prism through which the starting point of 103% debt-to-GDP should be analyzed. The document does recognize that the ratio is “too” high,[ix] and recommends policies meant to gradually lower the debt stock over the subsequent decade. But to get that ratio to bend toward sustainability, an enormous fiscal effort is recommended—thus the 3.8% of GDP primary surplus target. But for reasons I detailed above, this primary surplus target is a non-starter: Lebanon does not have the fiscal space for that kind of fiscal austerity.
The conclusion is that the magnitude of the debt operation assumed in the document is too modest. A larger principal reduction is needed. And, importantly, there is a need to marry any principal haircut reduction with even more aggressive maturity extensions and coupon reductions than those assumed by the plan. A debt operation is a traumatic experience and no country should go through it twice. It is infinitely better to go for aggressive debt operation—one that puts the country on a sustainable path—now rather than in 15 years.
To state the obvious, the more aggressive the debt write-down, the larger hit to banks’—and BdL’s—balance sheets and, as such, the more at-risk depositors are. That’s admittedly a difficult tradeoff to navigate. My own judgement is that the downside of a larger hit to the banking sector is less onerous, over the long run, than leaving the country saddled with a paralyzing debt overhang. But that’s a judgment that needs to be debated.
Finally, there is a conceptual omission in the plan that is relevant to the debt discussion. The document focuses on the government’s liabilities but ignores the asset side of its balance sheet. The amount of deleveraging that is needed is so large that one cannot off-hand reject usage of such assets. Any serious plan on state assets should thoughtfully consider equity considerations (who benefits and who doesn’t), modalities (outright sale versus securitization and better efficiency), and, most important, governance issue.
Foreign Exchange Market and Monetary Policy Regime
The document effectively acknowledges the de facto devaluation and recommends a unification of the multiple foreign exchange rates now in place under one weaker rate. After the unification, the plan recommends an eventual floatation of the foreign exchange regime while, reasonably, recommending very tight monetary policies to maintain LBP stability.
These are appropriate policy steps that I completely agree with. My issue, instead, is with the magnitude of the day-1 devaluation. I believe a larger devaluation is needed. Most analysts would agree that LBP was massively overvalued and that some adjustment was necessary.[x] And, appropriately, the plan foresees a large (59%) nominal depreciation over two years to LBP 3,684/$. However, it also projects high inflation over the same period. The result, though, is that the real depreciation is a significantly smaller 23%.
A currency is at equilibrium when both external and internal balances are regained. Start with the external side. The plan makes heroic assumptions on the Balance of Payments. It foresees the current account collapsing from $11.7 billion in 2019 to $2.4 billion in 2024. It also sees a dramatic improvement in the service accounts. However, basic calculations using standard trade elasticities would strongly suggest such current account savings are very hard to see with the assumed real exchange rate path. Further “expenditure switching” that renders imports less- and exports more-attractive is needed. That is another way of saying that the paper’s currency forecasts are too strong.
Turning to the internal side, there are three arguments worth considering when thinking about the “optimal” level of the foreign exchange. First, there is the growth angle. The document assesses Lebanon’s “potential growth” at 3%. I have issues with that assessment: The country has a sophisticated “tradable” sector—tourism being the most obvious example—that can propel it into a much faster pace of growth creating significant job-creating opportunities. A competitive currency will give the country the breathing space to grow faster.
Second, there is the medium-term inflation angle. If you start with an overvalued currency, you must accept a gradual weakening (i.e., back to equilibrium) over time. This, by definition, means persistent inflation and, in turn, would dictate tighter monetary policy than would otherwise be needed. Instead, a larger devaluation in year 1, while meaning higher than assumed inflation in year 2, will make the objective of low medium-term inflation more realistic and achievable.
Third, there is the debt angle. A larger devaluation automatically reduces the “real” value of LBP debt. While this admittedly constitutes a large wealth hit to creditors, it is a way through which the system achieves the necessary deleveraging that can put it back on a sustainable path.
To state the obvious, a larger day-1 devaluation does not come without cost. It erodes the purchasing power of wage earners who are already reeling from the recession. I fully recognize that this is a major issue with acute social and security consequences. However, I also see the issue as yet another trade-off. My own judgment is that the advantages of better competitiveness—external sustainability, faster growth, employment opportunities in the tradable sector, and debt inflation—outweigh the cost.
Some Concluding Thoughts: Gradualism versus Radicalism
To reiterate the point from the introduction, the plan is refreshingly serious. One cannot but applaud the government’s courage in adopting it. After years of meandering, there is finally a roadmap that can, at least, be the basis for a serious discussion on the way out of the crisis.
However, if there is one thread running through this essay, it is that the problem is enormous and that there is no easy way out. All policy levers are painful. Large as the assumed debt operation is, the country will still be left with a crippling debt overhang. Fiscal austerity of the magnitude needed to reduce the remaining debt is socio-economically a non-starter. The financial sector hole is so enormous that a large share of bank deposits is at risk. And, the magnitude of foreign exchange depreciation that is advisable is a major hit on wage earners, who can least afford it.
More fundamentally, the kind of wealth-destruction needed to rebalance the economy will create instantaneous impoverishment. Let me, for effect, repeat the math I started this essay with. The plan, as envisaged, effectively asks the society to halve its standard of living with no serious recovery anywhere in sight. Without wading too much into political-economic territory, one must worry about the implications of such economic collapse on a country with deep sectarian fissures and patterns of violent civil strife.
The above line of reasoning raises an important question. If policy levers are painful and seemingly socio-politically impossible to pursue, shouldn’t policymakers be reimagining an alternative plan? Isn’t there a space for a less-aggressive approach to crisis management—one that doesn’t try for perfection but can garner just enough political and social backing to avoid complete collapse?
What would such a plan look like? A detailed answer is probably best left for another essay. But in summary, it would be a plan that involves a less ambitious fiscal effort, that would focus more on revenues than expenditures; amortizes BdL losses over multiple years rather than clean it up in one go; compliments the deep debt haircut with even more aggressive maturity extensions and coupon reductions, while shifting, to the extent possible, the burden onto non-residents; and accepts a larger devaluation. Politically toxic as the topic is, this approach must rely, to some extent at least, on usage of state assets: The problem is too large for them not to constitute part of the solution.
A gradual plan will not avoid a deep recession but would limit its severity. The government will not avoid the IMF route, but it may mean a “smaller” ask—and lessor conditionality. Finally, and crucially, it means deposits will not have to be submitted to a haircut to the extent needed under a more aggressive approach.
Lest there be any confusion, I am not blind to the gradualist approach’s downsides. They’re major. The devaluation will be larger than under an ambitious program. The banking sector will remain weak and deposits will be frozen for significantly longer than is desirable.[xi] Capital controls will have to stay in place for years, and chances of a meaningful economic recovery will be low.
However, the gradualist approach has the major advantage of being socio-politically more palatable. In a complicated sectarian system such as Lebanon, the severe economic pain associated with a radical plan could well lead to non-linear scenarios including, at the limit, the return of the civil war. A gradualist approach would also give the nation time while waiting for oil and gas revenues and/or the possibility of a changed regional outlook. Most optimistically, it gives the civil society a few more years to organize and prepare to contest the next parliamentary elections with the hope that then the more profound transformation can begin.
If the perfect is unattainable, shouldn’t we strive for the good?

[i] To be fair, there were few serious attempts at a macro framework including the models used for the "A Ten Point Action Plan For Avoiding a Lost Decade" report. I would also highlight the excellent work by Jean Michel Saliba at Bank of America and by Alia Moubayed and Gerard Zouein. Worth also noting the rich twitter debate initiated by the NERDS that has produced thought-provoking insights and analysis.
[ii] A technical quibble—but one that can impact how short-term policy measures should be designed. My models suggest the country will face a deeper contraction in 2020-21 than that forecast by the plan. But, by contrast, my models also see a sharper recovery thereafter. Net/net, we end up in the same place but the journey to there is different.
[iii] The math on this is straightforward. I took the plan’s 2024 GDP ($30.4 billion) and compounded it at a potential growth rate of 3%. The level of GDP would return to the pre-crisis level ($59 billion) by 2043. I should add that I made the assumption that the real exchange rate does not change, meaning that inflation translates into an equivalent nominal depreciation.
[iv] Compared to Lebanon’s 21% of GDP, the average emerging economy spends 27.4% of GDP on non-interest government outlays. (Source: IMF World Economic Outlook data base). Admittedly, the amount of spending isn’t the best metric for public service effectiveness, but in standardized surveys of “government effectiveness”, including this one from the World Bank, Lebanon ranks extremely low.
[v] For example, the average emerging markets country collects 24% of GDP in revenues (Source: IMF World Economic Outlook data base).
[vi] It is regrettable that the document did not provide the underlying calculations behind those top line estimates. Nonetheless, the figures, broadly, are within the zip code of what others, including myself, have estimated for the losses.
[vii] This number was not explicitly stated but it had to be deduced. On page 29, the document states that the hole in the banks’ balance sheet is LBP 186 trillion. Within that total, there is the explicit assumption that banks will have to absorb BdL’s impairments which, from page 28, were estimated at LBP 121 trillion. As such, the document implies that banks’ own losses—those unrelated to BdL’s impairment—stood at LBP 65 trillion (i.e., the difference between 186 and 121). That figure, incidentally, is not too far from the numbers other independent analysts have calculated elsewhere.
[viii] It would have been desirable for the document to provide more transparency on the debt calculations. I tried to reverse engineer those myself. Using the document’s foreign exchange market and growth forecasts, and assuming $10 billion new additional debt (with IMF et al.), I come up with a 75% haircut on the face of existing debt.
[ix] If anything, and as the document admits, the starting point of debt will probably be higher than the assumed 103% of GDP: Intra-public sector debt—including that held by the social security agency—will be spared and the assumption of haircutting LBP debt at the same rate as Eurobonds is unrealistic given that that debt will have already been sharply reduced by the devaluation.
[x] See, for example, the latest IMF Article IV.
[xi] Without getting into the details this topic deserves, the gradualist approach can be complimented by segmenting the financial sector into “good” and “bad” banks. The former can start servicing the economy immediately and “funded” by small depositors who can be immediately freed. Legacy impaired assets—including claims on BdL—will reside in the “bad” bank and will be funded by large depositors. Those deposits will be frozen until BdL is recapitalized and the government debt operation is concluded.

Copyright © 2021 by the Lebanese Center for Policy Studies, Inc. All rights reserved. Design and developed by Polypod.