Frontier Informal Thoughts - Tackling Uncertainty in the Economy

In Detail

"The only function of economic forecasting is to make astrology look respectable." 

- John Kenneth Galbraith

Economics is a complex science and at many times, an obscure and often incorrect one. No further evidence is needed for this than Sri Lanka’s own story. Consider the hundreds of op-eds written in the local papers over the years, the thousands of hours of airtime given to commentators, and the millions of decisions taken in the lead-up to Sri Lanka’s economic crisis. Despite all this, it happened. Even with the benefit of hindsight, it is still sometimes a controversial topic without perfect consensus. 

Living in the middle of crisis, economics can feel like a daunting cloud of worries and lies all packaged together in a series of unending storms that never even rain properly. Even outside of crisis, economists often talk of some storm or the other, never letting the sun shine for too long. They throw numbers and equations at you, they jabber on about liquidities and deficits and above all, they never give a straight answer to a question.

“Give me a one-handed Economist. All my economists say 'on one hand...', then 'but on the other...”

- Harry Truman

This is even more infuriating when we look at the world we live in with so many unanswered questions that really matter. Who will be the president next year? What will next quarter’s energy tariffs be? Will interest rates keep going down? These are questions that any one would love to hear a straight answer on. Of course, economists aren’t fortunetellers, and yes, uncertainty is but a fact of life. But can’t they at least give a useful answer?

Amal’s success in setting out this framework for answering these questions was justified over the last few years. Starting from the constitutional crisis in 2018, and then moving across multiple momentous years - Easter Sunday attacks in 2019, Covid in 2020 and 2021, and the economic crisis since 2022 - Frontier’s team was able to take full ownership of talking about the wildly shifting economic landscape to our clients, and do so in a way that encapsulated the key values that Frontier built itself on - independence, trust, accessible communication, and constant change. Amal was of course, a key member of our team-driven process and his insights and contributions were highly valued. But it is a mark of his own success in building the team that he was able to be a part of the team instead of being the only member of it.

Since Amal’s passing, Frontier has continued the approach that we have been known for, and continue to engage with the economics in the same way. While our clients might have a good sense of the results of Frontier’s process, the actual process itself might not be as clear to everyone. In this memo, we’re setting out our research approach in detail so that the way we think, the way we come up with our research, and the way we engage with economics in a highly uncertain environment is laid out clearly for anyone to see.

Scenarios and ranges - Actioning an uncertain future

“We have to view the future not as an event which is predetermined and predictable or determined already, but as a range of possibilities”

​- Howard Marks

Long-standing clients would know that our approach to the problem of uncertainty has been to provide probabilistic range-bound scenarios for the future. Here we give our most-likely scenario on where a particular indicator can go, which we call our “planning range”. However, we also give alternative scenarios that can play out differently, which we call our “risk mitigation range”. At some points in the past, these have admittedly been quite complex (in 2009, we even tried to give a 3-D scenario!) but on the whole, we try to keep it between 3-4 relatively simple scenarios. These scenarios generally tend to have a few key factors that define them - for example, our current most likely scenario for the economy is that where the IMF programme continues with some level of delay and worries with nothing especially bad or good happening outside it.

Along with these scenarios, we generally prefer to avoid giving “single-number forecasts” and prefer to give “views ranges” instead. This allows us to account for the uncountable uncertainties that fall within our scenarios as well. Right now, even within our most likely scenarios, there can be varied smaller factors that change the specifics of what “IMF programme continuing” means. Particular policy measures, social responses, external events can all affect the economy WITHIN this scenario. We prefer to spend our time understanding the main factors that drive our scenarios as opposed to spending energy “dancing around a decimal” to try and perfect a specific number.

Despite this, there are instances where we have to give “single number forecasts” for certain practical needs of our clients. However, we recognize and communicate that the likelihood of these being precisely correct is very very low - and we don’t spend too much time trying to be perfect it. To try and account for the uncertainties, we do conduct regular reviews of such numbers and clients generally get at least a quarterly revised indication of the “forecasts”, but these are better understood as snapshot judgements as opposed to strong forecasts.

How is all this actionable? This combination of probabilistic scenarios and ranges in our view, gives a more accurate sense of what sort of outcomes are possible for the economy, even if it is not precise. We feel this allows for decisions to be taken that take into account all the realistic possibilities and not be overly affected if things don’t turn out the way we expect, and allows for the reality of uncertainty and incomplete information to be tackled nevertheless.

You can’t predict. You can prepare.”

- Howard Marks

Howard Marks’ quote that argues against predicting but for preparing, forms a core belief for Frontier. It has driven the way we give our views for two decades and will likely be the foundation of our thinking for decades more (though we are not predicting it will necessarily be so!). It is why we try to move away from using terms like “forecast” and prefer to use terms like “view” or “bet”.

In a 2018 podcast, Howard Marks further expanded on his view and it so well reflects our own thinking that we’re directly quoting it here.

“Now, it sounds like an oxymoron because how can you prepare if you can’t predict? But the answer is we never know what’s going to happen, but we can consider the likely scenarios and prepare for some of them. By definition, you can’t prepare for every eventuality.”

- Howard Marks

A recent example for this is our view on the LKR for 2022. We first released our end-2022 view for the LKR in August 2021, at a point where the CBSL rate of LKR 203/USD was allowed to be broken by the banks who had started quoting LKR 210-220/USD (though this turned out to be a small window only, and was soon reversed). Our most likely scenario at this time was that the LKR would end-2022 at around a 10% depreciation from that point, a view that ended up being wrong (though we did later update this view). However, we also had an adverse case scenario that had the exchange rate moving up to LKR 400/USD. 

What we told clients at this point was that although we felt a sharp depreciation was not in our most likely case, things were fragile enough that a very sharp depreciation beyond anything in Sri Lanka’s modern history was still a possibility. Thereby, our advice was to make sure that their business or investment portfolio was not going to be “knocked out” IF the currency DID depreciate so sharply. This would have meant limiting any USD liability exposure, considering what difficulties in changing pricing structures would have meant, and maintaining dollar and dollar-backed assets where realistic. Our most likely view ended up being completely wrong as the government chose to spend down its reserves from USD 4.1 bn in June 2021 down to near zero in early 2022 before approaching the IMF, but the adverse case scenario would have (hopefully!) served as a guide that could have still been useful for clients.

Frontier’s collaborative multi-faceted process - Our model is our team

“All models are wrong, but some are useful”

- George Box

Given that we take the uncertainty of the future as a core feature of our perspective, we rarely try to generate our view solely using theoretical, abstract, or quantitative models. How then, do we come up with our views and understanding of the economy? 

The process that Frontier has always followed has been that of a collaborative and multi-faceted approach towards a particular problem. That is, our model is our team and all their skills and application.

This process starts with team members looking at varied aspects of the economy (and increasingly commonly, outside the economy as well!) as part and parcel of engaging in macro intelligence. Economic data does form a big part of this “inquiry” but in most cases is only a single part of the question. While we don’t believe in using quantitative econometric models, we do still have a strong base in using as much data to substantiate our thinking as possible (our excels are often quite full and messy as a result!). However, our use of data is only a foundation from which we draw for our thinking. Varied other aspects, economic and otherwise, would be something we focus on in depth as well. 

​

A key part of this is how in Sri Lanka, the political and institutional framework allows and incentivises particular policy measures regardless of the economic “impacts” of it. That is, a critical question we spend a lot of time thinking about is “how do policymakers think?”. In some cases, we even find that thinking about this is even more important than the data and there are many instances (with the reserves drawdown from mid-2021 to early-2022 being the most costly example we can think of) where we have been reminded of the overwhelming power of this effect.

Beyond economic aspects, the team would also take in depth looks at other “non-economic” factors that could affect this. The last few years were the best example of these, as local and global factors outside of the domestic economy were key drivers of the story of the economy. Here as well, we take our wide-ranging approach that looks at data, analogous periods and events locally and domestically, expert opinion that drives expert behaviour, as well as calls on political decisions both by governments and oppositions.

"An economist who is only an economist is likely to become a nuisance if not a positive danger." 

- Friedrich von Hayek

​

An example from the present moment was how we thought about how varied factors could have affected the CBSL’s decision to continue to cut rates or not moving into 2024. At this stage in the economic and political cycle, the general trend is that of easing monetary policy, but how does the institutional impact of the IMF programme as well as the new CBSL Act affect these? How will the President and Parliament look at CBSL measures? How would businesses and investors respond to a cut? How would the probabilities of drought factor into these? How would instability in the Middle East affect these? How would the US election campaigns in 2024 affect narratives around the world?

Questions like these have very little “straightforward” answers, and definitely not answers that a model can give. In order to understand these, Frontier’s team would look at varied angles to try and understand the situation. Deep reading of the CBSL Act along with legal opinions, reading IMF reports and papers on the subject, looking at indepth reading of prior moments in Sri Lanka where factors were similar as well as comparatives from other countries, and of course a deep analysis of regular data would all serve in trying to answer this question. 

Simplifying complex and technical economics - Making it useful

"The master-economist must … understand symbols and speak in words. He must contemplate the particular in terms of the general, and touch abstract and concrete in the same flight of thought."

- John Maynard Keynes

A lot of the time, the economics we talk about tend to be complex and often driven by technical factors. Spending time regurgitating these technicalities is a relatively straightforward approach we can take, but not one that is particularly useful or understandable.

For Frontier, trying to make economics understandable and useful is a core way how we try to think of our work. In the past, when the complexities of the economy were less evident, we even termed this as “economics even your grandma can understand!” Since the pandemic and then the economic crisis, the complexities of the economy have magnified far more than we had seen in the past. This gives us a unique challenge - how do we both explain the complexities that matter far more than in the past but also not make it a boring academic data splurge?

One approach we have followed is the use of summaries, takeaways, and prioritization of our reports. Whenever we send out a report, we effectively split it into two - a short section with 3-5 key takeaways as bullet points, and then a longer structured section with more details. Even within the detailed section, we try to provide structured topics and questions, highlight important points, and provide quotes and bullets. This is particularly important in the context of more detailed topics, where most would prefer to get a top-line message, others would prefer to glance through topics, and perhaps only a few would want to do a detailed comb-through of the details we provide.

In especially detailed and technical topics - which have fortunately or unfortunately been more common as the crisis progressed - we run the risk of our economics not being particularly understandable to a grandma (unless the grandma has an economics degree). Our approach here is often to provide a series of Frequently Asked Questions as we run through the topic, starting from relatively introductory points before ending up on more technical points. Often, the key message is a simple “What does this mean for me?” - a phrase you would see quite regularly across most of our reports.

"Economics has become increasingly an arcane branch of mathematics rather than dealing with real economic problems." 

- Milton Friedman

​

Our use of charts and visuals are also part of how we think of our communication. This is true both in our reports and our presentations, where we follow a strategy of having just one point per chart to make it simpler to understand and often have relevant (though hopefully not excessive) annotations to explain the visual. Particularly in presentations, we do feel that our use of visuals, including cartoons and graphics, also helps to communicate messages without being overly swamped with numbers and technicalities.

Our constant engagement, particularly in our information curation products, is another way we think of engaging in easy to understand means. In fact, we internally like to think of our local daily and global newsletters as “macro-intelligence products”, since the consistent provision of simplified information about the economy is a critical way of how we think about the country and the world.

​

But how does all of this become useful beyond just information sharing? Whenever we can, we try to simplify the economics down to a key story - with relatively clear action points one way or the other. This often ties into our scenario thinking with our story generally related to our main scenario. For example, a  recent key story for us was that Sri Lanka was in a window of relative calm until early 2024 - and this was driven by a relatively technical set of points on domestic debt, inflation statistics, weather events, and the timing of elections. All these underlying factors aside, we felt that the story and framing of a “window of calm” results in an easy basis to think of action points -  that the window was a relatively short period of time for either restructuring, long term planning, or even short term opportunities. We feel that providing these stories can also help make the underlying economics easier to grasp and more importantly, easier to act upon.

Avoiding bias - how we try to maintain our independence

"It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" 

- Upton Sinclair

One of the most important values that Frontier holds is our independence. Clients know and trust that the views and economic research that Frontier provides are based on our understanding of the economics from an unbiased perspective, and that we have no agenda that we’re trying to push that could affect the independence of our views. This trust we have cultivated comes from years of consistently demonstrating this, but its something we don’t just passively give lip service to. Frontier takes its independence very seriously and actively moves towards ensuring it. The overall way we try to do this is by trying to identify specific incentives and biases that could cause us to lose this trust and actively moving to distance ourselves from these.

A key way we ensure this is by having a strong and clear refusal to take up any work that is tied to government and especially partisan political parties. Our view is that any engagement with such can easily end up with incentives, suasion, and even restrictions on what we can say and what we can’t say.  By strongly avoiding such work, we ensure that no political or partisan agendas seep into our work and we are able to stay neutral.

This ties into our refusal to make any public policy recommendations. We feel that by providing such policy recommendations, we are liable to become tied to them, both directly and indirectly. It becomes harder to provide an unbiased independent view when we have been advocating for one policy or the other - since our view of the future is going to be clouded by whether our advocated policy happens or not. For example, if we advocate for a specific tax proposal, our view on the fiscal performance of the government might become affected by whether the government takes our proposal or not.

Along with this, we try hard not to limit ourselves to any particular economic ideology or viewpoint. This is not to say we believe in all of them, but rather that we’re able to both take insights from varied sources but also be in a position to explain these more objectively as needed. A good example of this was Sri Lanka’s experiment with MMT in 2020 and 2021 - by taking a neutral look at the topic, we felt that we were able to give a more realistic perspective of what it could look like - short term growth, asset market booms, and eventual inflation if not balanced by sharp taxation. If we took a particularly supportive view of MMT, we might have argued that inflation was not a consequence of the policy while a particularly opposing view of MMT could have missed out on the short-term results of MMT before inflation took over.

"In economics, the majority is always wrong." 

- John Kenneth Galbraith

​

Although as a firm we can strongly avoid such political, policy, and ideological stances, it is not something that we can fully avoid as individuals. All members of our team will have their own political, policy, and ideological views and that will undoubtedly affect the way they think about the economy in our “team model”. While we are unable to prevent this from happening, we do take strong steps to limit how much it can affect our work. We take a two-fold approach here.

First, we encourage and sometimes require team members to disclose their personal incentives as much as possible, and we effectively consider that these biases could affect their work directly and indirectly. By having constant and open communication about these, we believe the team is able to to limit the impact of their personal views onto their work. The diversity of views at Frontier also helps limit how much a single perspective can dominate as well.

Second, we take explicit measures to look at the counter argument to any work we do. Taking after a military technique, we call this our “red team” process. Whenever our work leads us to a particular conclusion, we spend a lot of time trying to disprove ourselves, to find alternative information, and sometimes even “run the numbers” again from scratch with different assumptions. Often, this results in a more nuanced view that might directly affect our most likely message or at least is part of our alternative scenarios.

​

Another key incentive we stay away from is our refusal to move towards handling or managing any funds. This is an obvious incentive that can cause us to give overly positive messages due to a fiduciary responsibility, or at least avoid specific topics given how they can affect markets (and thereby, any hypothetical funds under our control). By avoiding this, we’re able to be more neutral in our views and clients can be fully aware that we’re not affected by an incentive to protect a particular pool of money.

Beyond this, similar to many firms that work in the broad field of finance and economics, we also have general restrictions on market trading. However, even outside of this, we make it a point to be extremely open and transparent within the team on what sort of financial incentives affect individual team members and their views, beyond the biases we discussed earlier. By being open and transparent about these internally, we’re able to recognize the biases that affect us and then consider if they may be clouding our views.

​

These factors often end up with us being quite contrarian to both “traditional economists” but also to the market’s view. In many cases, we are able to argue that a particular policy direction can have positive impacts in the short-term even if we might agree on their long term unsustainability. On the flip side, we also won’t shy away from pointing out the consequences the economy could face beyond the short-term even if the short-term is quite positive.

​

Act fast and course correct - dealing with a rapidly changing economic environment

"When the facts change, I change my mind. What do you do, sir?" 

​- John Maynard Keynes

The last few years have particularly shown us the importance of recognizing volatility and rapid changes that occur in the economy. In many cases, these can result in completely different trajectories to the economy and thereby to what decisions are prudent. While this is obviously true for both the impacts of our economic views, it is also a critical part of giving the views themselves. 

We usually try to keep our views consistent for some time, in order to make them more actionable as well. The uncertainty and rapid movements are thereby baked into the ranges and probabiltiies we give, and we usually try to explain changes in the economic environment through the application of these views.

However, the recent economic story shows that sometimes the very frameworks of the economy can change incredibly rapidly. The move from a low rate environment to a high rate environment for example, isn’t just on the rates but on the entire economy and the back-and-forth from it that can affect the entire story completely differently. In many cases, there is also very little information on what can happen, what can cause changes, and sometimes on what can change itself.

“Understanding how to act under conditions of incomplete information is the highest and most urgent human pursuit”

- Nassim Nicholas Taleb

A key part of our approach here is to incorporate ranges and probabilities to account for a wide variety of possibilities, but also to rapidly change our views as required. The challenge here is to do so in a proactive manner that reflects genuine changes in circumstances and evolving understanding of the economic situation, and to avoid a reactive change to specific smaller economic events. The line between the two can be blurry at times, but often we try our best not to be overly distracted by the volatility of the moment but still make changes as required.

An example of when we feel we got this right was in late 2021 and early 2022. From August 2021 to January 2022, we changed our view on the USD/LKR rate 3 times, as more and more evidence materialized of Sri Lanka rapidly using up its foreign currency reserves much faster than we expected. From January to June of 2022, our ranges stayed consistent, but we updated our probabilities to reflect the period of massive economic and political upheaval the country was facing.

Moments like these where the speed of economic change is so evident will be rare. However, we still think Sri Lanka is in a place where uncertainty and volatility are a big part of the economic story. To account for this, we’re now moving to a semi-formal end quarter review of our views. The ideal outcome of these reviews will be to reaffirm our views, so that the consistency of a view can be maintained without being reactive. However, the regular review then allows us to course correct if needed and be proactive in our changes.

​

Where to from here?

​

As Sri Lanka moves through quite a momentous period in its economic history, the importance of understanding the economic pathway is undoubtedly a core part of engaging with Sri Lanka. As the political pathway ahead shows, this is not just an economic story but a deeply diverse one, that takes perspectives and alongside many other factors as well. With all of these, it's clear to us that uncertainty and worry can easily remain as core feautres of the Sri lankan economic story.

​

Frontier has always placed itself squarely within these uncertainties, and we see our role in the economy as clarifying as much of these uncertainties as possible with independent research and providing actionable frameworks to navigate the uncertainty as well. We don't doubt the future will be uncertain, but we stand ready to navigate it alongside you all.

​

You can’t predict. You can prepare.”

- Howard Marks

​