MORGAN STANLEY / Brazil - How Much Lower Can Rates Go?
arthur garbayo
How Much Lower Can Rates Go?
February 01, 2012
By Gray Newman | New York & Arthur Carvalho | Sao Paulo
Gray Newman: Arthur, with Carnaval less than a month away, a lot of investors I speak with believe that Brazilians have a new reason to celebrate. Just this past week Brazil's central bank stated that there was an "elevated probability" that the target Selic interest rate was on its way to single-digits. For over a decade now we've been waiting to see single-digit interest rates in Brazil and they've never materialised except in 2009 (and early 2010) when we were in the midst of a global recession.
I don't mean to put you on the spot, but you don't seem to be joining in the celebration. You have the Selic reaching 10% this year. Why the caution? Did I misread the minutes?
Arthur Carvalho: Thanks Gray. Today, we're adjusting our forecast for the Selic interest rate to reach 9.75%, down from our previous forecast of 10%. But you are right; even with our new forecast, we are maintaining a more cautious stance.
I think there are a couple of issues here. First, I agree that the minutes from last week are fairly clear: Brazil's central bank would like to bring interest rates to single-digits. But my job is to forecast what I believe that the central bank will end up doing, not what it wants to do. You can argue that some members of the central bank's board would like to do much more in cutting rates, but I think there is a limit as to how far they will go. And I suspect we are already nearly there.
Second, remember that we've seen time and time again that the central bank has often followed up a statement that was viewed by the market as too hawkish with a statement that was a bit more dovish. If you remember in the last week of December, many read the quarterly Inflation Report as very hawkish. The language from last week's minutes may in part be in response to this.
Gray Newman: Arthur, I understand your second point about the central bank wanting to fine-tune its message and the need it has to balance out perceptions when the market starts to move too far in one direction or another. But the language in the minutes was pretty clear. The central bank seems to be on its way to a 9% handle - a view I've been highlighting for a long time - or even lower.
Arthur Carvalho: I'll give you that. While I was looking for a dovish set of minutes to offset the more hawkish tone of the Inflation Report, I was surprised as well as just how dovish the minutes sounded. But that brings me back to my first point: my job is to determine what the central bank will do, not what it would like to do. And on that front, it's worth re-looking at Brazil's inflation dynamic
Year-on-year inflation should slow in the coming months due to base effects and to some extent during the whole year due to the new weights, given the components of the IPCA measure of inflation. You will recall that food inflation was heavily influenced last year by supply shocks that are being replaced by much more benign readings this year. This should take inflation from the current 6.5% to 5.3% or even slightly lower depending on the harvest, although the weather of late has not been cooperating - we've been suffering from a new bout of drought. But listen, I'm not a meteorologist and could be surprised.
But after this base effect plays out, I expect that inflation will stall around 5.3% to 5.5%, and this should end the downward trend that has been giving the false impression that Brazil's inflation problem is over.
Meanwhile on the growth front - which seems to be the other implicit goal from the central bank - things look significantly better. The epicenter of softer growth was in 3Q and was largely concentrated in the industrial sector that was undergoing an inventory correction that led to flat GDP reading in that quarter. Although demand softened compared with the pace seen in early 2011 (due to tightening measures), it was still stronger than supply. I'd argue that the resilience on the demand side can be attributed to the labor market strength we've been highlighting (see "Brazil: The Cyclical and Structural Inflation Dilemma", This Week in Latin America, January 23, 2012). Unemployment is now at record-low levels - it reached 4.7% in December - and real wages that looked softer in 3Qr were up 2.6% in real terms in December, hardly a worrisome scenario for domestic demand.
I have no doubt that the central bank would like to take the Selic to single-digits, what central bank would not? But the central bank's goal is to ‘control' inflation, not simply reduce the interest rate and then face a greater inflation problem later.
I am expecting another 50bp cut to 10% at the March meeting, followed by a last 25bp cut in April, given that now it will be politically costly to take back the single-digit rates promise. But the notion that in late May the central bank can continue to cut when faced with stronger growth data and stable inflation well above the 4.5% target seems unlikely even for our currently dovish central bank.
And remember Gray, just a few weeks ago, I was being criticized for being too dovish. If you look at the DI curve in January right before the central bank announced the last 50bp cut, it was pricing in the final cut to 10.25% while I was forecasting that the authorities were on their way lower still to 10%.
Gray Newman: Fair enough Arthur. We've seen the market swing from one end to the other on the rates call in Brazil. I remember this time last year when you were arguing that the market was misdiagnosing Brazil's inflation problem. Rather than focus on the underlying uptrend in service prices, many had fallen into an easy extrapolation from a handful of beef and produce shocks to headline inflation. That easy extrapolation then collapsed.
And despite my first question about why you had only modestly adjusted your 10% forecast for the Selic rate to 9.75%, I actually agree that the real issue here is not whether the central bank goes to 9.75% or to the 9% range.
The way I see it, Brazil has been facing two challenges. The first is a structural challenge of how to live with a strong currency which is damaging Brazil's industrial base while boosting consumption (the Growth Mismatch). There is nothing wrong with counter-cyclical policy, but the Growth Mismatch is a structural issue that is unlikely to be resolved by adopting counter-cyclical policy.
And the second challenge is a cyclical one seen most intensely during 3Q11, which was primarily concentrated in weakening output as demand from abroad suffered. Again there is nothing wrong with engaging in counter-cyclical policy, but in Brazil's case it appears that the shortfall was more a matter of output suffering due to external factors than because domestic demand had collapsed.
Arthur Carvalho: Indeed, here's what I've been telling investors since late last year: Be prepared for the Growth Mismatch to reemerge in 2012. That was a harder sell in November than it is today. At the time, I had to argue that Brazil would see a rebound in stronger domestic demand in 2012 based on the policy response.
Although many have doubts about how effective monetary policy is in Brazil, I am actually a believer. The tightening cycle did slow down credit in 2011. More importantly, macro-prudential measures shifted credit growth towards overdraft and credit cards, rather than auto and personal loans. And now the central bank is reversing itself on most of this; right now we are on the exact opposite side of the early 2011 story. Monetary policy is as loose as it has been in the past 15 years. It takes time, but this stimulus should be kicking in later in 1Q. I am hearing everywhere from auto sales picking up to banks saying that they are already passing through the softer credit standards. It worked to slow down consumer demand, and it should now work to boost demand as well.
Also, do not forget the impact that the minimum wage will have in consumption: not only is the labor market in a multi-decade tight level, but in February the minimum wage paid will also be 14.1% higher than the month before. We hear estimates that 44 million Brazilians earn a minimum wage. And remember that more than half of those - around 25 million - are pensioners with little or no marginal propensity to save. Yes, we should hear a fiscal announcement over the next few weeks with significant budget cuts, but even if the authorities announce an R$60 billion cut, expenditure will grow above 10% this year. And last but not least, the development bank, BNDES, has just received a new round of fresh capital to continue to expand its balance sheet, and we can already see evidence of a pick-up on BNDES loan expansion in December's credit data. All this will be much more evident in 2Q, exactly when the single-digit Selic is set to appear.
Gray Newman: As you know Arthur, whenever you use the phrase the Growth Mismatch, you've got me on board. But let me play devil's advocate here for a minute. While our US economics team is still working through the tension between FOMC members' projections for the fed funds rate and the official FOMC statement, it remains of the view that the Fed is paving the way for QE3. It's easy to imagine a world in which the Fed and the ECB remain very accommodative for a long time, which, in turn, could put pressure on commodity prices higher and the Brazilian real to strengthen further.
Can't you argue that Brazil's central bank is simply trying to stave off excessive inflows to Brazil with its more dovish tone? After all, the Brazilian central bank hardly appears to be alone. The Chilean central bank surprised Luis Arcentales earlier in January with a rate cut. Luis has long argued the Chileans would ease rates, just not in January, given stronger-than-expected inflation prints. And if you read the January 20 communiqué from Mexico's central bank, it was about as dovish as you could imagine without cutting interest rates. After having warned of FX pass-through in December, Banco de Mexico now downplayed it as a risk. In fact, Banco de Mexico went so far as to remove the specific reference to the exchange rate as a factor that needed to be "closely monitored" and repeated its line that the next move could be an easing, given more accommodative stances of central banks in the developed world.
Arthur Carvalho: I have no doubt that the authorities are concerned about rising commodity prices and inflows to Brazil. The exchange rate seems to be crucial to the new industrial policy that aims to save embattled domestic-oriented industries. Clearly, every time the Fed and ECB signal further quantitative easing, these generate concerns, but not all concerns lead to the conclusion that rates should be cut further. Remember in 2010 a big part of the inflows from foreign capital was actually being channelled to consumer credit, given that banks were raising capital abroad and expanding their balance sheets domestically. And higher commodities prices can also be inflationary. So, to imagine that the central bank will simply react by cutting interest rates aggressively could complicate things further for Brazil. Indeed, I don't think lower rates will be the primary tool to avoid currency appreciation.
Gray Newman: Just to be clear here Arthur, the central bank is going to have a difficult time now backtracking on the suggestion that Brazil is heading to a single-digit Selic target. Once you are that explicit, it is hard to stop short.
Is part of the intent of the minutes to slow down the gains of the Brazilian real? Can you imagine a much stronger real prompting interest rate cuts to the 8% range or lower? What could come next?
Arthur Carvalho: I think that a much more dovish central bank could lead to a weaker currency, but in order to actually end the carry trade, the authorities would have to reduce rates much more aggressively than my target of 9.75% or your talk of rates closer to 9%. Are we on our way to 8% or lower? I don't think so. This would in turn create too much stimulus for domestic demand and would result in an even further acceleration in services prices and eventually in headline inflation. The administration has been quite clear when it comes to dealing with inflows and currency appreciation. I would not be surprised if it hikes even further the IOF in derivatives, which it perceives as the most important channel for real strengthening.
Gray Newman: OK, I think I understand you. You have interest rates moving lower for three reasons: first, that is what the central bank has signaled; second, the base effect is allowing for a temporary downward trend in the year-over-year numbers; and third, the softness in the economy - largely due to an external shock - has been enough to keep expectations largely in check. But that puts you in a fairly uncomfortable position. You don't have inflation returning to the 4.5% target anytime soon, certainly not in 2012. Doesn't that raise the risks that inflation could move higher along with expectations and that the authorities could be forced to hike interest rates in 2013?
Arthur Carvalho: Although we are forecasting a much more benign headline inflation path that will be below 5.5% for a good part of the year, core inflation will likely continue to be above 6.0% and raises questions about the appropriateness of the policy stance. But I don't expect the central bank to reverse course and hike in 2012. Headline inflation should be lower due to a series of one-off effects such as energy prices revision and no bus fare hikes due to municipal elections. On top of that, I am assuming that food prices will behave well. Of course, if there is any supply shock on the food chain, we could have a headline number accelerating very sharply, and it could get above 6% in a couple months. Let's hope that the possible QE that our European and US teams are expecting doesn't have the same impact as it did on commodity prices in 2010.
Finally, I believe that growth was behind most of the aggressive easing cycle in 2011: many of those fears should be tamed as we see stronger data on the activity front. I believe that by April there will be enough data to prompt a smaller 25bp cut to 9.75%, although not enough to signal the end of the cycle. But by late May, I believe that it will be increasingly hard to justify continuing to stimulate an economy with a significantly better growth outlook knowing that monetary policy lags from six to nine months.
But I do think that a hike in interest rates will be necessary in 2013: we're looking for rates to move back up to 11%. You'll notice that the hiking cycle in 2013 is smaller in part because I suspect that the authorities will opt for a mixed approach using both macro-prudential measures as well as the Selic. Again, all of this is assuming that the world continues to muddle through. If this is the case, I expect that the Brazilian economy will continue to accelerate through 2012, and this should continue to pressure services prices. As you mentioned, this will most likely de-anchor inflation expectations and pose a risk to headline inflation in 2013.
Gray Newman: I cannot help but worry a bit here about two things in 2012, long before we get to 2013. First, we could be setting ourselves up for Growth Mismatch in overdrive. Imagine a return to strong inflows pressuring the currency stronger and, in turn, boosting the purchasing power of Brazilian consumers once again. The bout of currency weakness last year appears - at least in the balance of payments data - to have helped to ease some of the spending and trips taken abroad by Brazilians. But that could all come back with another bout of currency strength. Despite all the measures taken in recent years, the Brazilian real managed to gain dramatically.
Second, I still worry about opportunistic policy-making. I keep hearing voices in Brazil calling for the need to re-set interest rates much lower. The proponents of this policy recognize the importance of monetary policy and will admit that there is a place to engage in monetary policy tightening as well as easing, but believe that the starting point of Brazilian interest rates is fundamentally wrong. And so when last year's global slowdown appeared and threatened to morph into a full-blown global recession, they argued that it was the right opportunity for the central bank to cut rates sharply. A full-blown global recession didn't play out, but in three or six months' time we could be faced with QE3-induced strong inflows, and this might be another argument in favor of a much bigger cut to interest rates. As you said, what is important is not what the central bank intends to do; our job is to forecast what it will end up doing.
And I highlight all of this because whatever you think of where monetary policy should go, I think you will have to agree with me that among the greatest challenges for Brazil's long-term growth story, adjusting monetary policy is no where near the top of the list.
Arthur Carvalho: I completely agree on that. But that is another discussion for another time. I've got to get ready for Carnaval after all. While we don't wear costumes in Bahia, I am planning on a special version of our Growth Mismatch t-shirt to mark the debut of the Morgan Stanley Latin America Economics team's participation in the festivities.
Para ter mais informação acesse estes portais: www.desenvolvimentistas.com.br e http://www.joserobertoafonso.ecn.br/