Model this taper and show your work, if only verbally

Shortly following the announcement of a delayed taper:

The rupee rallied 2.8 percent to 63.4950 per dollar, according to prices from local banks compiled by Bloomberg. Thailand’s baht rose 1.2 percent to 31.850, the Philippine peso gained 1.4 percent to 43.87 and Malaysia’s ringgit appreciated 1.2 percent to 3.29. Global funds pumped $5.7 billion into the stock markets of India, Indonesia, the Philippines, South Korea, Taiwan and Thailand this week, according to exchange data.

Pay special heed to quantitative magnitudes.  For how long are we delaying the taper?  One or two months?  How much is the taper anyway, relative to the stock of relevant financial assets?  Taking $10 to $15 billion off of $85 billion a month in purchases, when the asset stocks are in the trillions?  Woo hoo.

Here are some interesting comments from Stephen Jen.

I’ll say it again: none of you understand what is going on here, and neither do I.  I am not seeing enough admission of this basic fact.

Addendum: Scott Sumner offers a response.


I’ll say it again: none of you understand what is going on here, and neither do I. I am not seeing enough admission of this basic fact.

You are talking about the whole macroeconomics, I hope.

Look, macroeconomics was too difficult for Keynes, and he was a genuinely clever chap.

Macroeconomics is the high brow equivalent of technical analysis. Ultimately nonsense, but sounds (or can be made to sound) plausible enough for people to make careers out of it.

True, exchange rate forecasting is very tough. I once saw a blurb in a Economics (Barron's Business Review) that mentioned the factors that influence exchange rates, and they were over a dozen. Non-linear system, so at best all you can do is perhaps forecast a long-term (steady state) relative minimum or maximum value.That's why playing the Fx is a zero-sum game as any investor will tell you.

Know your branches of economics!

That graph should have agreed upon by economists on the y axis and listened to by politicians on the x axis.

It's sometimes a stretch to imagine there's something even to be understood here.

All we see is a complex response of multiple interacting systems. Why can't we accept that and end of story. Why the perennial hunt for an elusive grand model.

I would have argued that it was a signal of overall fed dovishness, which could have market impact beyond the mere taper. However, the 5 year yield collapsed both in nominal 5 year treasuries and TIPS, with breakeven inflation generally unchanged. The fed announcement, then, decreased real rates of return by ~30 bps. What model allows the fed to move real rates without using inflation expectations as a lever?? My only conclusion is WTF, it must be magic.

I believe market monetarists, actually, think monetary expansion with no increase in inflation expectations is perfectly consistent. Monetary expansion increases NGDP expectations, which could come in the form of RGDP or inflation or both. But I'm not sure whether they would have predicted that this announcement would not move TIPS.


Inflation expectations don't have to change - It's just markets reacting to a risk-on/risk-off trade.

And that's TC's mistake, I think. Sure, risk-on/risk-off is kinda self-referencing but, in this case, with a clear starting point (The Fed decision), I think it makes sense.

The fact that the real-world effect or even the impact on financial flows is small doesn't mean that markets can't overshoot... Again, that's kind of easy/lazy explanation but it doesn't mean it's untrue and it doesn't mean, ex-ante, that it was the only possible outcome, if somehow markets had chosen to interpret things differently...

My theory, sorry, no way to show work:

Institutional investors have been collectively attributing stock market performance to QE for so long, that they have been conditioned to treat it as a signal for asset prices. Personally I think the impact is overstated, but it doesn't matter as long as enough investors cause asset flows because of it. More QE = buy, less QE = sell, timing and magnitude almost irrelevant.

It's true that we don't know what's going on in the real economy because of all this, but it's easy enough to understand what drives fund flows.

Agree. All that matters is the meme that got loose among investors. Perception becomes the reality in market related matters.

I'm a buyer of your theory.

'I don't get it; therefore no one can get it' is a frequently dangerous, if occasionally accurate, position.

On the other hand, "You people are all idiots" is a frequently dangerous and frequently accurate position.

My theory- the market has one more piece of evidence that there will be no taper.

My theory:

The Fed is propelling money in a widening gyre
The emerging markets cannot hear the Fed;
Things fall apart; the taper cannot hold;
Mere anarchy is loosed upon the world,
The blood-dimmed bubbles are loosed, and everywhere
The emergence of BRICs is drowned;
The best bloggers lack all conviction, while the worst
are full of passionate intensity.

Trick question. "Never reason from a price change."

Never reason from a price change in isolation. However, to use the price response to a well-specified event as a starting point for reasoning about economic system properties is completely legit.

Monetary base is about 3600 billion currently. So 10 billion for 4 months is a bit over 1% of monetary base. Which corresponded to a roughly 1% drop in the relative value of the usd. This doesn't seem odd to me in the least.

QE is a risk-on policy. Tapering is incrementally risk-off. The surprise move to hold QE at current levels was therefore a relative move to risk-on. Emerging markets = risk-on. Flows to emerging markets will pressure those currencies upwards. When Bernanke hinted at tapering in June you saw the reverse as capital fled and EM currencies took a hit. Furthermore, interest rate differentials in EM countries still attract flows and the real rates in the U.S. just dropped on that Wednesday news from the Fed. We can quibble over how this all seems short-term and optimistic given maybe a few more months of $10-$15B of security purchases, but nonetheless, the above at least is what I think I've seen in actual markets.

My thoughts exactly.

My model is that the market believes the difference is a matter of 2 years not 2 months. Will show my work when I get home from the Cuban festival. The short version is this: the market is right. Actually, that is also my long version. Back to the bongos.

What about Ryan Avent's conjecture: Do you really think we (and global markets) have only been learning and reacting to the start date of tapering?

Also your last statement is vacuous ... and incorrect, I hear a lot of questions being asked and possibilities considered. It is called unconventional policy for a reason, right?

Think of it as a feedback loop where what the Fed does affects what the Fed does. That's why 2 years not 2 months.

It's about the signal. World markets taking comfort along in the fact that the Fed is back on track with reality.

A Number of developed market currency jumped hugely on the move. Cherry picking certain developing market currencies frames the argument where the observer specifically wants it.

Take Sumners advice and just watch equity mkts

I am not entirely sure that the falls in the Indian and Indonesian Rps is not due to local investors getting nervous about the inherent structural defects in those countries, slowing growth through effects other than the QE taper. But taking the falls as linked to the threat to start withdrawing QE, it is pretty obvious that there is concern that the Fed does not get it, and there is thus a risk that early tightening is possible. Once the Fed issues a more dovish statement, investors are now more re-assured that the Fed does get it, and will not tighten early. So we are not talking about the effects on markets of the purchases of bonds in the last month, but the view of markets on Fed policies in the forthcoming years. More broadly, they are looking at the consensus of policy makers in the US and in the rest of the developed world. The US Fed would not be issuing statements that the rest of the US and developed world policy makers would consider outrageous, it just doesn't work like that (you could call it the group think effect if you are being unkind, or consensus if you are being nice) . So if the Fed can issue re-assuring dovish noises, that means the other policy makers are also likely to be at least neutral, if not favorable to this idea. So in essence the initial announcement about tapering was an announcement of higher interest real rates in the world, which impacts high growth countries first. The latest series of announcements indicate somewhat lower real interest rates can be expected.

I am not sure how much this is about expectations that these currencies will rise in value as opposed to expectations that the U.S. dollar will lose value. It is relative and it is about the U.S., not countries in Asia.

On a related note, I am not sure that we should be paying attention to what happens in India during the last week. That is too short-term thinking. The longer-term trend, over the last year, is that the Indian economy has run into some speed bumps. What happened during the last wee and what will happen next week are irrelevant.

IF the spread between EM returns and "recovery mode" US returns are very narrow, would that create a kind of knife edge with capital jumping back and forth as policy hesitates to advantage either one?

Alternatively, it's a

What about some game theory? The Fed is out of "real" ammunition. What they've got left is an awesome ability to bluff everybody. The market can't "call their hand" collectively to find out what a crappy hand they have. Optimistically, this is Nick Rowe's Chuck Norris in action. However, I wonder how long the game can go on. I haven't heard anybody muttering about what kind of situation the Fed would be in if a new recession started, say sometime next year and GDP was shrinking. Could they still stay in the game and bullshit everybody?

"Taking $10 to $15 billion off of $85 billion a month in purchases, when the asset stocks are in the trillions? Woo hoo."

Marginal vs. average. As the very name of this blog suggests, it's the marginal effect of something that's significant for decision-making and for assessing effects. So, sure, I have no trouble believing that a marginal change can have a noticeable effect.

The fed has managed to recapitalize the financial system, and the fools that would have jumped out of their windows late 2008 are still at it, and have managed to build an even more fragile structure of debt and leverage. A small change and everything goes to hell.

What would all these assets be worth without the fed? The monthly purchases are close to the monthly increase in the treasury debt, enough to set the price. Maybe in fact the us is a banana republic.

Sumner nails it: “The real reason why surprises are so important is that they cause market participants to revise their expectations of the future path of policy”

Expectations uber alles. The Fed has essentially absolute control over the currency, the only question is what they choose to do. The markets are saying "the Fed is less stupid (or more pro-growth, if you prefer) than we thought."

This is another of those situations where market monetarism so, so clearly explains things best. How can anyone seriously argue the Fed is powerless when things like this happen?

It also bears repeating in the same vein: the actual size and scope of QE is almost irrelevant, the real issue is what QE says about the expectations of what the Fed will do in the future.

(Of course, the Fed could avoid having to do QE at all if it would just change the target, but I'm already a broken record on that subject.)

That's exactly how I see it and where I place my money.

Ah, you have grasped the point. The way to make money in the market is to follow the Fed's direction. Anyone who has tried to short the Fed has lost. Their power is not because they have power, but because the market follows their direction.

That credibility cost how many trillions to repurchase after 2008?

It's significant that investors know that there is historically a Gargantuan serial correlation between one Fed move and the next. So "tapering" is interpreted, not without reason, as tightening for several years to come. The reversal of that language is interpreted as continued loose policy for several years to come.

My theory is that the serial correlation is explained, as I say above, by a positive feedback loop in which whatever the parent tells the child the child assumes is the new policy and then the parent, without necessarily realizing it, acquiesces to the new policy the child now takes for granted.

Even if my theory about the feedback loop is wrong the serial correlation in Fed policy remains as fact.

I'm no economist obviously but just an ignorant individual speculator whose market return has been above average, at least according to the IRS.

Economists need to think more like traders. The best traders, as far as I can tell, do not have static models. The market gets in different configurations depending upon, I don't know, whim, randomness, chaos? Speculators try to follow the temporary order that is made out of the chaos without losing sleep over what explains the new order.

As Victor Niederhoffer likes to point out: it is like the changing cycles at a horse track. One day word gets out that people are making money on long shots, everyone crowds into long shots and that is the day to be betting on the favorite.

Markets are about fads, not classic truths. No static model will ever explain them.

Niederhoffer, isn't he the guy who lost like $1,000,000,000 in about 30 seconds?

"off of"? You surely know "off" is a preposition and does not use another preposition.

Tom, Is this statement gramatically correct? "Tom is a jerk off of the rarest sort". Sorry, couldn't resist.

In June the market was surprised by Bernanke's taper-talk which didn't reflect reality. It viewed the talk as a tightening of policy which was political and not backed up by the data. The non-taper this month surprised the market in the opposite manner with the view that the Fed is returning to reality. Bernanke was kind of snippy at the press converence and said the nontaper was a result of the data and there was no regime change, but markets thought otherwise. It simply could be that the Fed is playing it safe because of the probable government shutdown but the market doesn't see it that way. The market may be seeing the nontaper as a precursor to the new regime come January. That regime won't end QE without the data to justify doing it.

"none of you understand what is going on here, and neither do I."

Animal spirits.

Remember the repo market. Perhaps the real channel is in the possibility that the globalized repo market has largely returned to using Treasuries for collateral? The value of that collateral is the PRICE of bonds, not the interest rate. Any expectation that the price will drop (as would happen if the Fed stopped buying) would throw a big wrench into the massive overnight borrowing. Indeed perhaps a massive wrench, Econ Crisis II. Under this theory, basically: Fed easing is not about injecting money, it is about keeping up the price of bonds, i.e., keeping the shadow banks liquid for overnight borrowing, in order to unwind their toxic junk, and to keep wheeling/dealing within the financial sector. (More or less entirely within the financial sector. Because it's not like non-financial businesses need it: they are sitting on $2 trill cash and demand is still slack...) Emerging markets would then be nice little side bets for the financials, especially if U.S. interest rates are not going up.

I have a model, but it isn't mathematical. (These days, I would aver, math may hobble you: You need more compartments than can be computed.) The repo market is modeled intermittently in these two videos:

Comments for this post are closed