Colombia can cut big, but there are ifs
Colombia has been dipping its toes into the rate-cutting game. There is room for a lot more. But inflation must continue to fall, else room for cuts evaporates. While we are comfortable that inflation does fall, we find even given that, there is not huge value in longer dated spreads to the US. Nice pick-ups, but future drop-offs maybe not as nice
Mexico can cut big, Brazil less. Now what about Colombia?
Recently we sent out a short report showing the Banxico has built up a buffer to cut its key rate by 2.7% based off a simple a steady state analysis. It’s here. That has since moved to 2.5%, reflecting a moderate rise in Mexican inflation. We did the same for Brazil, and showed that from here on out, cuts from BCB really need to be bolstered by further falls in Brazilian inflation and/or cuts from the US Federal Reserve. It’s here.
We now ask the same question of Banco de la República de Colombia (BanRep).
We use the same three ingredients to help identify room for policy rate manoeuvre:
- The external nominal official rate differential is the spread that the markets focus on, and it’s the one that directs FX forwards.
- The internal real rate is the domestic policy rate less inflation – the real policy rate. It’s the one that affects domestic circumstances.
- The external real rate differential is the domestic real policy rate versus the real Fed funds rate. It’s the link to control of the currency, and feedback to domestic inflation.
Inputs to the calculation process for the official rate and market rates
Note that the market data are based off respective swap rates (not government bonds)
For Colombia:
- The classic rate differential is 7.4%, versus a 4.5% average. The delta is 3.0%.
- The internal rate is 4.4%, versus a 0.7% average. The delta is 3.7%.
- The external real rate differential is 2.2%, versus and 2.1% average. The delta is 0.1%.
Applying an equal weighting to all three identified elements gives us 2.2% (see below).
Here’s how we calculate rate cut potential for BanRep based off a steady state
The issue here is the external real rate differential, which is practically flat to the historical average currently. Should BanRep cut by 220bp it would push the external rate differential to zero, and the delta versus average to -2.2%. We don't mind deviations below average for the identified spreads, but not by more than one standard deviation. We find that one standard deviation from average would be 1.7%, so we limit the cut potential to that.
Bottom line, cut potential with a reasonable degree of comfort is to 11% (from 12.75% currently). If so, the resulting external real rate differential of 50bp feels tight, but bear in mind that this spread has traded negative on quite a few occasions in the past decade and a half. Also, while this is a steady state analysis, it really needs some reasonable expectation for future falls in inflation as a support. We can't tolerate an opposite outcome of higher inflation.
But, let’s take things one step forward.
Hypothesize the following, 1. Colombian inflation falls by 200bp (to 6.3%), 2. The Fed cuts by 2.25%, and US inflation falls by 1%. Then we’d have the following:
Adding some future dynamics – lower inflation and Fed cuts. What then?
Now we’d have, 1. The classic rate differential at 3.0% above average, 2. The internal rate at 3.5% above average, and 3. The external real rate differential at 1.1% above average. Now an interest rate gap has re-opened for BanRep.
Applying equal weighting gives us 2.5% for that gap. So, BanRep can ultimately chop its official rate to 8.00%. That's calculated at the current official rate of 12.75%, less the additional 2.5% identified here, less the 2.2% gap identified above (including an add back of the prior 50bp breach of the standard deviation in the external rate differential).
And, to push this to the extreme, given all of the above, plus let's assume Colombian inflation hits the top of the inflation range (4%), then the capacity is there to get the official rate down to 6.5%. Now, the stars would really have to allign to get to this level.
Market swap rates are already pricing a lot – so actually not that much room for more
And what does all this mean for swap rates? Currently the 10yr spread to the US is 3.5% and the 2yr spread is 3.9%. These are well through the historical averages (back to 2008).
The 10yr spread to the US at 3.5% is just 50bp above the historical low at 3%. And history shows the 2yr has capacity to the down side to a sub-2% spread. Then again, should the Fed get to 3% and BanRep threatens to get to 6.5% that gives a 350bp official rate spread, making the achievement of a sub-200bp 2yr spread to the US far from easy. Or ar least it would require a path to an even lower BanRep rate.
Based off this it's tough to see conviction value in Colombian spreads to the US given the amount of cutting that would need to be delivered to secure meaningful performance. We do, however, see value in dis-inversion strategies; negative in carry, but likely to perform.
And what about the FX angle? USD/COP is now in the 4,000 area (slightly through) and the forwards see it at 4,225 on an 18-month view. That’s the degree of COP depreciation that should (theoretically) be tolerated by BanRep. The long run averages as identified set floors below which FX vulnerability can arise. Currently USD/COP looks a tad stretched to the downside relative to the cumulative inflation experience, presenting some vulnerability for COP. That said, the ING view for USD weakness on Fed cuts offers a meaningful buffer. See more on that here.
If the stars align, BanRep can get to 6.25%, but they really would have too align well
BanRep has already cut by 25bp two times, from 13.25% to 12.75%. That’s been facilitated by a material fall in inflation from 13.4% to 8.4%. Inflation remains too high. But the fact that it is falling and expected to continue to do so helps frame the analysis. We identify room for BanRep to cut by more. We think by a cumulative 170bp, while not deviating too much from historical averages. That would bring the official rate to 11%.
Down the track if we assume US inflation of 2%, a Fed funds rate of 3% and Colombian inflation of 4% then there is a path to 6.5%. But, a lot has to go right to bring that about. And already there is a lot priced out curve. We find 2yr and 10yr spreads to the US optically attractive from a pick-up perspective, but not especially from a performance one. The 10yr spread for example is just 50bp off it’s historical low (back to 2008). There is more relative value in Mexico.
We’ll cover Chile in detail next – stay tuned!
Download
Download opinion
Padhraic Garvey, CFA
Padhraic Garvey is the Regional Head of Research, Americas. He's based in New York. His brief spans both developed and emerging markets and he specialises in global rates and macro relative value. He worked for Cambridge Econometrics and ABN Amro before joining ING. He holds a Masters degree in Economics from University College Dublin and is a CFA charterholder.
Padhraic Garvey, CFA
This publication has been prepared by ING solely for information purposes irrespective of a particular user's means, financial situation or investment objectives. The information does not constitute investment recommendation, and nor is it investment, legal or tax advice or an offer or solicitation to purchase or sell any financial instrument. Read more