Update on scenarios after SVB failure

Ok, we now know what the player next in turn did: The regulators and government officials decided to take out the bazooka right away without hesitation. As I said yesterday, "this time around this would come much sooner than it did in 2007-2008 – the action during pandemic showed that with large financial crisis in recent memory, the point when political establishment acts is significantly earlier than otherwise.". This is especially so given that Yellen is sitting at the Treasury.

This changes the likehood of the scenarios quite a bit:

Nothing happens: 20-25% >>> 30-35%

Tremors without casualties: 25-30% >>> 30-35%

Some casualties, but situation remains contained: 25-30% >>> 10-15%

Contagion which is eventually contained: 10-15% >>> 5-10%

Nothing happens: 2-5% >>> 2-5%

Effectively, with forceful action from the government providing backstop to the uninsured depositors the middle of the probability distribution has hollowed out: It is much more likely that this will end as soon as it started. However, this does not mean that adverse outcomes are ruled out. It is just that now they would require other mechanisms to occur, which is less likely. This even applies to the middle outcomes: I dont think we have a good model of depositor's beliefs formation, and it is still likely that at least some of them will act out of caution despite the government guarantee. It is just that now it is rather unlikely this will reach a critical mass. 

Rather than the depositors, the behavior to watch now is of bond holders, sources wholesale funding and stockholders. Bond and stock holders were wiped out in SVB, something that officials did only in rare occasions in 2008 out of fear. It remains plausible that they will pull out in more meaningful fashion. And there is still possibility that there are vulnerabilities we do not know about. Therefore, the probability of truly adverse outcomes has not decreased much.

Addendum: Looking at my probabilities, I realized that the wording of "without casualties" is not aligned with the exact idea in my head. I really mean no meaningful casualties. E.g. few small banks failing still qualifies as "no casualties". By no casualties I mean no banks in the size range of SVB or larger fail.

Scenarios after SVB failure

 

The first thing to realize that at this moment failure of SVB is a failure of single financial institution. And while SVB was not a small bank - with north of 200 billion in assets it ranked as 16^th largest in the US – neither it is a large-enough bank for it to be systematically important. To realize this, notice that it felt under both the more strict (250 billion) and even the more relaxed (500 billion) threshold for participation in annual Federal reserve stress test exercise. Therefore, in isolation, its failure does not pose a threat to the financial system. The question, though, is what does this imply for the rest of the financial system.

Of course, this is impossible to know with certainty. In such situation, it is useful to spell out possible scenarios, and try to assign (necessarily subjective) possibilities to each scenario.

Nothing happens

Option 1 is that this will be fully self-contained to SVB, and then after few days financial markets realizing the limited nature of the situation will move on. There are many reasons for this case. First, SVB was clearly unique among large(r) banks in terms of its vulnerability. On asset side, it had unhedged interest rate exposure, while other large banks have their interest rate exposure hedged away, something that is a public knowledge. On liability side, SVB was funded by uninsured short to much bigger degree than other banks, partly due to its rapid growth, as its assets quadrupled since start of the pandemic. Moreover, the depositors were from relatively tightknit community, which meant that the news about weak position of the bank spread very quickly among its depositors, with some of the voices of the community, such as Peter Thiel acting as amplifiers. We know that such conditions make run by depositors much more likely.

In other words, SVB was uniquely vulnerable among large(r) banks, both on asset and liability side. And this uniqueness might mean that its failure will be a one-off and no other banks – or at least no other banks with any meaningful size – face pressures and fail. There are ample precedence for isolated failure of non-systemic bank in situation when other banks do not share its vulnerabilities on asset and liability side.

Likelihood: 20-25%

Tremors without casualties

While SVB was unique among large(r) banks, it does not mean that many of the small and medium banks do not face similar problems. Over last couple months there were news of this bank segment in the US facing funding pressures related to rising interest rates and associated outflow of funds, as depositors opted for higher yielding government money markets.

Moreover, whether SVB was unique or not really is subjective. It is entirely plausible that depositors will consider SVB more similar to other banks, especially Californian banks. Indeed, there are signs of troubles at few other medium-sized banks with similar profile as SVB. This could mean that small or large share of depositors decide to pull their money from these banks.

What happens after that really depends on the scale of fund outflow. If it is relatively small, possibly limited by appropriate noises from government and regulatory officials, then this pressure could eventually die without claiming any other casualties. This is especially likely if the resolution of SVB is favorable to its depositors.

Likelihood: 25-30%

Some casualties, but situation remains contained

On the other hand, if the resolution of SVB is unfavorable to uninsured depositors – or only eventually favorable, with doubts about their fate in coming days/weeks – then the likelihood that the pressure on other small and medium banks yields further casualties is large. However, this does not mean that things will get out of hand. It is very plausible that after few bank failures the  situations calms down, as financial markets and depositors realize that the problems are limited to few institutions rather than widespread. Or alternatively, after some period regulators and government officials come up with a forceful action. In essence, this scenario amounts to weeding out the weakest banks without ramifications for the rest of the financial system, something that we have also seen before. This is in line with academic literature that suggests that it is the weaker banks that typically fail.

This and previous are the most likely scenarios: for things to get out of hand and us to end up with more or less limited financial crisis, it is not enough for one bank to fail (as long as it is not central to the financial system). You need the wider financial system to face the same vulnerability, either on asset or on liability side, or both. While there are some shared vulnerabilities on asset side, as far as we know, these rather limited. And on liability side the wider banking system is in very different situation, with large amounts of excess reserves. And on top of everything, the large systematic banks are very well capitalized, as far as we know.

Likelihood: 25-30%

Contagion which is eventually contained

While direct implications from SVB failure is limited, and its vulnerabilities are not shared by the wider financial system, one cannot rule out that the wider financial system will be affected. The most likely reason for this to happen is for the resolution of SVB to be unfavorable both to bond holders and uninsured depositors., This could plausibly be because of limitations on interventions from regulators and the Fed imposed by the new regulatory rules after the global financial crisis. This could sawn doubt into bond holders and depositors about the backstop offered to other institutions in case they get into trouble. As a result, they could pull to safety of either other instruments or larger, safer institutions. And like that the crisis would start to spread to wider financial system.

In such an event it is likely that an action to stop the contagion would eventually come. The typical script of financial crisis is gradual encroachment of the crisis onto wider and wider circle institutions until the pain becomes too large for the political system and there is resolute action. Plausibly, this time around this would come much sooner than it did in 2007-2008 – the action during pandemic showed that with large financial crisis in recent memory, the point when political establishment acts is significantly earlier than otherwise.

Likelihood: 10-15%

Gradual escalation to full blown financial crisis

Despite all this, things could get evolve into full blown financial crisis. Financial systems always seem resilient until they are not. It is possible that the problems of SVB are more widely shared, or that the repercussions of its failure will reveal vulnerability we do not know of. If the global financial crisis taught us something it is that there are parts of financial system which are little known until they blow up. In the same way, we might be sitting in front of ticking bomb and not know about. While plausible, this remains rather unlikely.  

Moreover, we are clearly in early stages of developments, so even if we were to go there, it would take at least couple months of gradually worsening situation. The analogy of SVB to the 2007-2008 period is the early failures of summer 2007, and especially the Northern Rock episode. Northern Rock was actually very similar in many ways: the rapid growth prior to failure, the reliance on short-maturity liabilities, and only limited problems on asset side. Both SVB and Northern Rock were vulnerable because of their funding structure rather than because of the asset side of their balance sheet. If this analogy is suggesting something, then it is that even if this will become something serious, we would need to go through the gradual escalation, that would take months, if not longer.

 Likelihood: 2-5%

Macroeconomic implications

If SVB will prove to be a standalone, then the story will be soon confined to historical textbooks and will not have any ramifications on wider macroeconomy, beyond likely making Federal reserve more cautions and opting for 25bps hike on Wednesday, rather than 50bps hike which was the baseline as of Thursday morning.

If it leads to pressure on other banks but no other failures, then the implications would be very similar, albeit it might lead to tightening of financial conditions and weakening of the economy sometime down the road. The implications for immediate monetary policy would be again limited, albeit the hiking might stop one or two meetings sooner then otherwise.

This is more likely under the scenario where there will be other failures, but things will not get out of hand. Depending on the speed of developments, this could mean that March hike is the last one by the Fed, while ECB will very likely make at least one more hike in this scenario.

What if this continues to gradually escalate? Assuming that things don’t worsen rapidly it would likely mean recession later this year and corresponding rate cuts. This means that only rapid escalation, which remains rather unlikely, could mean that the ECB would stop hiking already in May. Rather, this puts into question the hike in June and July, and has implications for size of hike in May.

 

 

 

 

 

 

Is it or isn’t it a recession?

On regular basis we are now having discussion in Europe whether “something” is or is not recession. Inevitably, these discussions lead to conclusion that while we started talking about the aspects of that “something”, the discussion is really about “what is recession” rather that whether this particular something is a recession. In other words, the discussion we need to have is what constitutes a recession, rather then discussing the details of the economic situation.

Czechia example

Consider the example of Czechia. Here are the facts:

• GDP has contracted for two quarters in a row, each time around 0.3% non-annualized. It is still below its pre-pandemic peak.
• Consumption has contracted for 5 quarters in a row, cumulatively 7.6%.
• Fixed investment has decreased in last quarter as well, albeit after strong recovery throughout the previous year and a half.
• The reason why GDP did not drop more is because net exports surged from their extremely low values reached during the pandemic period. In last quarter government consumption also helped a lot.
• Despite all the weakness, labor market is tight, with unemployment rate close to its pre-pandemic historical lows (in case you don’t know, it is ridiculously-sounding low at 2.1%).

Is this a recession? In many dimensions it is. I mean, 7% drop in consumption is simply gigantic; for comparison, during great recession, there was almost no contraction. Now it is accompanied also by drop in investment. And then there are the two successive declines in GDP, which would qualify as recession according to the silly definition for “technical recession”. On the other hand, the 0.6% drop in GDP is far-cry of normal recession. And above all, it just does not feel like recession here on the ground. It just feels like pretty depressing period for consumers.

Are there recessions in RBC models?

Yesterday, my boss had an interesting idea. Maybe we should look at things like defaults and bankruptcies, rather just macro aggregates, when determining whether something is or is not recession. I then built upon that idea further and to put focus on self-reinforcing demand shocks as the key aspect of recessions.

To see this start with following question: Are there recession in RBC models? The question of recessions was one of the first controversies between proponents and opponents of RBC models. The reason was that RBC models explained fluctuations in output as fluctuations in the productive capacity of the economy, rather than output falling below such productive capacity. This led to the classical critique question of “What are the productivity shocks”, as declines in productive capacity required negative productivity shocks. Such shocks were at odds with the understanding of productivity at the time, given that productivity was understood as akin to technology or knowledge. (By the time I was reading this discussions I was not that puzzled: if you think about productivity more broadly it feels very natural that it could easily decrease following a shock; an example below).

Therefore, RBC models did and did not feature recessions. They did in the sense of output declining in some periods of time. But they did not, in the sense of output being below its (short-run) potential. I.e. there were no output gaps. Only the move from RBCs to new Keynesian models allowed for this possibility more broadly - albeit for a long time only to a limited extend – by introducing nominal rigidities which could give rise to demand driving the decline in output. And I think this contrast in perspectives is actually very useful for current situation.

Real shocks and demand shocks

To see how the perspective of real shocks vs demand shocks is useful now, consider back the case of Czechia. How can we explain the current economic developments? I think it I fair to say that the economy is undergoing large real shocks, both positive and negative. On the negative side, the surge in energy costs is a very large and - crucially – real shock, in that it changes the productive capacity of the economy. Moreover, at this point in time it also leads to redistribution of output away from consumption, which to some degree might be optimal. Meanwhile, the reason why output did not collapse more is because at the same time the economy also faced large positive real shock in form of global supply bottlenecks unwinding, which in turns I just reversal of previous large negative real shock. (All of these are examples of negative productivity shocks, since productivity and technology are separate).

What is crucial, though, is the fact that these real shocks have so far not been accompanied by negative demand shocks. In the words, the typical mechanism though which recessions are self-reinforcing – drop in demand causing further drop in demand through various channels – has not kicked in. This can be seen in things like firms not firing people or aggressively cutting down their orders and slashing inventories in response to uncertain demand environment. It can be seen in that the financial channels being so far very silent: there hasn’t been a wave of defaults leading banks to curtail the supply of credit. Simply, while both consumers and firms are being hit by shocks, they have not started worrying about the outlook in a way that would cause them to change their behavior above and beyond what the shocks require.

No recession – but what is it then?

And this is why I don’t think Czechia is in recession. Neither is any other European country so far. Simply, a recession really requires a demand shocks to occur, and they have not occurred yet. I mean this in the sense that demand is so far only responding to the real shocks hitting the economy, and is not additional source of shocks, at least not to a significant degree.

Of course, what is not yet the case might soon come to be: there is nothing that says that firms will not lose their faith as their order books are depleted and start slashing their demand and laying of people. And if that starts happening, then it could gain speed pretty quickly. But in absence of this happening, the economy will proceed along its current trajectory, struggling to generate much growth before the effects of real shocks start to reverse. Will the resulting path be a recession? I don’t think so. At the same time I am sure it will not be fun either. Seems like we need a better terminology.

P.S.:  Was pandemic recession a recession in this view? I would say yes, because there was a demand component. Financial channels were very alive and kicking. Firms like car producers slashed they demand for inputs expecting slow demand from their customers. Sure, the demand was not the dominant factor, which is why the recovery from the recession was very different from great recession. Of course, the main reason why demand played limited role is because of the demand management: fiscal and monetary policy prevented collapse in demand and actually gave additional boost to demand. But that is another story.

Can we enter recession without layoffs?

On recent episode of Inside Economics podcast, there was a question raised whether we can get recession without layoffs. There are two parts to this question: Can we enter recession without layoffs, and can we go through recession without layoffs? I think the latter is pretty hard to achieve and it would require very mild recession, at which point it is a question whether given period actually amounted to a recession or not. It the former question which is more interesting.

So, can we enter recession without layoffs? I think the answer is not only “Yes”, but that we have already had such experience. Consider a following pair of pictures capturing labor market aggregates (first picture) and GDP with unemployment rate (second picture) in particular historical period. (On purpose I omitted the date axis, which will be revealed at the end, but bonus points for guessing it outright 😉) .

The second half of this period was officially categorized as recession. This is not very surprising as during this time monthly hiring has declined significantly from 5.2m to 4.5m, dragging down with it the employment, which has declined by 1.7m from 138.4m to 136.7m. Meanwhile, unemployment rate increased from 4.7% to 6.1%, while GDP had two quarters of negative growth intersected by a positive quarter. Overall, this is clearly consistent with recession, albeit not a particularly severe one.

The key point, though, is in the layoffs series, replicated once more with slightly extended sample:

While layoffs did increase a bit, the increase was very small amounting to 200k-300k, less than half of decline in hirings. Moreover, this increase occurred only towards the end of our sample. Before layoffs did not record any increase (at best they were bit elevated) while employment already managed to decline by 0.5m.

The conclusion is simple: we have already had an experience when we entered recession without widespread layoffs. Therefore, even if firms will hesitate to lay off people – something that I think will be a feature of potential 2023 recession – we might enter recession in 2023. All that it would take would is for firms to significantly decrease their hiring, something that is possible even if they will be hesitant to let their current employees go. In such situation employment will start to decrease and unemployment rate will start to increase, eventually tipping us into recession. And then the layoffs will likely come. (Indeed, one can imagine that if we do tip into recession, then the eventual layoffs might become more aggressive then initially thought, given that firms will be postponing layoffs for long period of time.)

So what was the period of time captured in the pictures? It was the onset of Great Recession:

Of course, after the end of the picture employment continued to tank as hiring decreased further, and more importantly, layoffs increase:

The other way to see the argument is to look at the cumulative effect of each of the flow category on the overall employment:

This shows that hires were initially the predominant source of shortfall in employment and that layoffs really kick in only in late 2008. Hence the conclusion that high layoffs are not necessary for unemployment to rise and for us to enter recession, low hiring will suffice. That said, given that this picture is quite sensitive to the choice of base date – layoffs were somewhat elevated in November 2007 already relative to say 2006 - a more qualified statement would be that large-scale layoffs are not necessary, low hiring and slightly elevated layoffs will do.

This is because even in normal month there is large amount of churn, with lot of laid-off people who quickly find new work, as long as hiring is robust. If it is not, then people who will be laid off as part of natural process, will not find work and will become unemployed.

P.S.: Someone might argue that this whole period might not have been classified as recession unless the financial crises escalated in September 2008 (official declaration came only in December 2008). That is possible, even if not very persuasive, given that employment decline was gathering pace already in the summer. 

Addendum to ‘The power of choice of transformation, part 3’

When I was going through my rant about why year-over-year growth rate is not a good transformation for this period of volatile macroeconomic data, I might have made it look that I oppose it only during this period. In reality, I quite dislike year-over-year growth rate most of the time.

My biggest pet peeve is how is it used in news to indicate whether something has contracted or not in a given quarter. Take, for example, the Chinese GDP during 2022, shown in terms of level (blue) and in terms of growth rate:

 

It is obvious that Chinese GDP experienced two quarter when in contracted, that is second and fourth quarter of 2022, dropping by 2.7% and 1.1%, respectively in quarter-on-quarter terms (not shown here). Yet, in both quarters the year-over-year growth rate remained positive, effectively thanks to the accumulated growth in previous quarters.

And then comes a newspaper writer and writes “Chinese economy avoided contraction in second  quarter of 2022”. Come on, how useful and correct is that?! What really annoys me is the terminology "in given quarter here". There is no way to characterize the experience as avoiding contraction in given quarter, because within that given quarter there was a contraction. The absence of contraction in terms of year-over-year growth rate is about the other 3 quarters...

But oh well, I know that this just me insisting on words having a proper meaning...

The power of choice of transformation, part 3

 

The last part in this series is dedicated to the Covid recession, which  by the fact that it was lead to hugely anomalous data points makes life of macroeconomic analysts much harder. Here I will discuss how especially the usual transformation used in economic print, the year-over-year growth rate, end up not communicating almost anything to the user. In other words, this piece is really not about demonstrating power of transformation, but rather the lack of power.

Background: Basic transformations

Bit of background. Economists are mostly interested in how are things changing, rather than how things are. For stable macroeconomic time series like unemployment rate this poses no problem: graphing level will show you both where we are, and whether we are going up or down. But of trending time series this is not the case – if you take long enough sample then the increasing level screws up with the scale so that all you see you are now higher than in beginning, but not whether you have recently increased or decreased.[1]

The obvious solution is to use growth rate. But which one? We have simple growth rate, annualized growth rate, year-over-year growth rate… and recently even things like year-over-two-years growth rate! All of them have their place in the macroeconomists’ toolkit, and it is up to the economist to choose which one to use, giving us the power of choice of transformation. That is why they pay us the big bucks. Just kidding.

The most popular transformation in economic press is the year-over-year. This is probably for two reasons for this choice. First, year-over-year gives you comparable numbers like annual growth rate.[2] This is also true for annualized growth rate, which explains its use. However, year-over-year growth rate also has the advantage of not being too volatile, which is not the case for simple growth rate and especially not for annualized growth rate. This for example explains why year-over-year is THE statistic used for reporting GDP growth rates for emerging markets, which have more volatile quarterly series than developed countries, which either use simple growth rate (in case of euro zone) or annualized growth rate (in case of US).[3]

YoY growth rate and pandemic

This then explains why people usually use it. But here comes the catch: year-over-year growth rate is sometimes a horrible statistic for telling you the story. To see why, consider the case of euro zone GDP. Everybody roughly knows the story: GDP decline in 2020q1, then collapsed in 2020q2, before partially rebounding next quarter. Afterword it was mostly flat for next two quarters, before rebounding during the second and third quarters of 2021.

Now consider looking on the year-over-year growth rate of euro zone GDP during the pandemic:

 

Can you immediately tell the story from it? Well, for starters, there was a huge collapse in 2020q2. So far so good. But what about the  development in following quarters? Well, year-over-year growth rate was still very negative in 2020q3-2021q1, albeit less. This tells you that GDP was still lower than year before. Finally, in 2021q2 there is a spike, followed by a drop.

Looking at this you would hardly conclude that by far the biggest rebound was in 2020q3, unless of course you spend quite some time thinking it through. Instead, you would conclude that it was 2021q2 which was the best quarter. Of course, this was not the case and the strong year-over-year growth rate was really about the low comparison base from year ago – that is 2020q2.[4] Similarly, you would conclude that 2021q3 was much worse than 2021q2, which is not true. Simply, in periods of abnormal movements, such as the pandemic, year-over-year growth rate will often tell you more about what happened during this quarter last year, rather than the current quarter.

The lost power of transformation

But this drawback of year-over-year growth rate is not the main point of this post. Rather, the point is that this chart fails completely in it the main goal expected from charts: telling a story to the reader. Or even worse, it tells a misleading story. And yet, this is a chart which was common during the pandemic, and is still common. Simple, the authors are throwing away their power to tell a story with a chart by choosing the wrong transformation.

P.S.: What is the solution? Of course, simple growth rate would be the best growth rate here. But personally, I became a huge fan of indexed charts. Why? For starters, they tell you the same story like level of the series, which is really the story to tell here – down a lot, then partially up but not all the way.

 

 

Simple growth rate will struggle to communicate this – and might be even misleading due to its non-linearity. Second, unlike levels, index tells you also relative magnitudes. In the original level graph you could see that GDP went from 2,600,000 to roughly 2,300,000. But this is a completely useless information in that if it would be instead 5,200,00 and 4,600,000 it would tell you the same story. Simply, most readers have no idea about current level of GDP and such information is not valuable to them. Worse, what is really relevant to them is by how much are we lower, e.g. by 10% or 20%? To figure this out they would need to do calculation in their head.

And here comes in the power of index charts. From index chart you can see that you went to 85, and hence 15% below the index  quarter. Moreover, you can also easily read off the magnitudes of quarterly movements to get growth rate: for example, going from 85 to 96 tells you that you increased by 11% of the pre-crash value.[5] Hence index chart tells you the level story, while also containing basic information about the story of quarter-to-quarter changes.

P.P.S.: Lately I also started using index values for analyzing sequence of monthly data. Namely, I index all data to the same month of last year. Like that I can see the year-over-year growth rate, if I want to, but also have the story of levels, undisturbed by the base effects, which would be the case if I would be showing year-over-year growth rates.

 

 

 

 

 

 

 

---

[1] Unless of course you are looking at Italian GDP. 😉

[2] By annual growth rate I just mean the simple growth rate for data in annual frequency.

[3] Fun story: back during the depth of pandemic Czech business daily had article comparing the performance of US, euro zone and China. The problem? It was comparing the numbers are reported by statistical offices, and hence comparing simple growth rate with annualized growth rate with year-over-year growth rate. Yes, that is sometimes the quality of Czech business journalists.

[4] While it was a good quarter, that was just a coincidence which is almost irrelevant to the shape of the curve: Year-over-year growth rate would be huge even if 2021q2 would show moderate decline.

[5] Note that this is not the same as growth rate, which was more like 14%. That said, I would argue that it is more valuable information than growth rate: you don’t really care that you increased by 14% from 2020q2 level; it is more valuable to know that you have increased by 11% of the pre-crash value.

The power of choice of transformation, part 2

The second piece in this series is dedicated to modelling and to monetary policy. It has been motivated by this nice chart from the Goldman Sachs:

 

This chart is highlighting that we should not think that recession is coming just because of the tightening we have seen already, because lot of the effect of tightening has already been felt simply because tightening has started almost a year ago. Which I think is important point to make and the chart is very useful in putting numbers to this idea.

That said, after thinking about this chart I started wondering whether it does not overstate its case. The odd thing that stroked me was the fact that the effect of the tightening is tending to zero over time. At first glance it does sound reasonable that after tightening ends the effect it has should gradually become zero. And in-so-far as focusing on the process of tightening (that is raising rates), this is just logical conclusion.

However, that is not how this graph is being used and interpreted. The chart was being used to argue that the impact of monetary policy as such is going to wane in coming quarters, which is a different point from the impact of just the tightening process. The point to realize is that the end point of tightening is relevant; in other words, if we end up with high interest rates[1], then that should have negative effect on growth irrespective of whether the rates are being raised further or not. In the parlance of Fed, they want to raise rates to rates to restrictive level and keep them there. The use of word “restrictive” implies that that high level will have continuously negative effect on aggregate demand as long as rates are at that level. In other words, the effect of tightening will be non-zero even after the end of the tightening process; as higher rates stay, the effect will stay as well.

How is this all related to power of transformation? Well, in the report where the chart was used the authors come clean and say that the model links the effect on growth to changes in the financial conditions index. This then leads to the conclusion that when changes in monetary policy go to zero, the effect goes to zero. However, the choice of changes in financial conditions as the driver of growth is not as self-evident as it might seem. I can as plausibly argue that it is the level of financial conditions that is relevant for growth, with high level implying lower growth. Indeed, this is what one would take away from standard macroeconomics models.

Ultimately, the point is not that either changes or level is the correct choice.[2] I think it is fair to say that both matter: level is important, but rapid changes in level can carry much bigger punch than the change in level alone would imply, and hence changes are important too. The point is different: The story the chart (and underlying model) say is really a function of the choice of transformation; if different transformation would be used, the story would be completely different. Hence the power of choice of transformation.

---

[1] High of course needs a benchmark, as in high relative to what. Here I mean high relative to (long-term) neutral rates.

[2] I would view this view irrespective of econometric arguments. For example, I am pretty sure the authors chose changes in financial conditions index either because they  concluded that the series has a unit root, or because they concluded that it fits better. Neither of these would persuade me that this is the only correct perspective.