Archive
The Important Trade Effects Are Longer-Term
The question about the impact of trade wars is really two questions, and I suppose they get conflated a lot these days. First, there’s the near-term market impact; second is the longer-term price/growth impact.
The near-term market impact is interesting. When the market is in a bad mood (forgive the anthropomorphization), then trade frictions are simply an excuse to sell – both stocks and bonds, but mostly stocks. When the market is feeling cheerful, and especially around earnings season, trade wars get interpreted as having very narrow effects on certain companies and consequently there is no large market impact. That is what seems to have happened over the last few weeks – although trade conflicts are escalating and having very concentrated effects in some cases (including on markets, such as in commodities, where they really oughtn’t), it hasn’t dampened the mood of the overall market.
In fact, the risk in these circumstances is that a “happy” market will take any sign of a reduction in these tensions as broadly bullish. So you get concentrated selloffs in single names that don’t affect the market as a whole, and when there’s any sign of thawing you see a sharp market rally. We saw a bit of that in the last day or so as Mexico’s President-elect and US President Trump both expressed optimism about a ‘quick’ NAFTA deal. Honestly, the broad market risk to trade in the near-term is probably upwards, since any increase in tensions will have a minor and concentrated effect while any thawing (especially with China) at all will cause a rally.
But beware in case the mood changes!
As an inflation guy, I’m far more interested in the longer-term impact. And there, the impact is unambiguous and bad. I’ve written about this in the past, in detail (see this article, which is probably my best on the subject and first appeared in our private quarterly), but the salient point is that you don’t need a trade war to get worse inflation outcomes than we have seen in the last 20 years. You only need for progress on advancing global trade to stop. And it seems as though it has.
Not all of the forces pressing on inflation right now are bullish, although most are. Apartment rents have slowed their ascent, and the delayed effect of the dollar’s rally will have a dampening effect next year (arrayed against that, however, are the specific effects of tariffs on particular goods) although globally, FX movements are roughly zero-sum in terms of global inflation. Money growth has slowed, to levels that would tend to contain inflation if velocity were also to remain stable at all-time lows. But velocity recently started to uptick (we will find out on Friday if this uptick continued in Q2) and as interest rates gradually increase around the globe money velocity should also quicken. The chart below (source: Enduring Investments) shows our proprietary model for money velocity.
At this point, trade is pushing inflation higher in two ways. The first is that arresting the multi-decade trend towards more-open markets and more-numerous trade agreements fundamentally changes the inflation/growth tradeoff that central banks globally will face. Rather than having a following wind that made monetary policy relatively simple (although policymakers still found a way to louse it up, potentially beyond repair, largely as a result of believing their own fables about the powerful role that central banks played in saving the world first from inflation, and then from deflation), there will be a headwind that will make monetary much more difficult – more like the 1960s and 1970s than the 1980s, 1990s, and 2000s. The second way that trade conflict is pushing inflation higher is mechanical, by causing higher prices of recorded goods as a direct result of tariff implementation.
But this second way, while it gets all the ink and causes near-term knee-jerk effects in markets, is much less important in the long run.
Tariffs Do Not Cause Price Declines
Adding to a good’s price does not make its price decline.
It’s worth repeating that a couple of times, because it seems to be getting lost in the discussion about tariffs – in particular, in the discussion about tariffs levied on US commodities. Grains prices have been plummeting, as the chart below showing front corn and soybean prices (source: Bloomberg) illustrates.
There are many reasons that grains prices may be declining, but if “tariffs have been levied on US production” is one of them then there is some really weird economics happening. Corn and soybeans are commodities. Specifically, this means that they are essentially fungible – corn from site “A” is essentially the same as corn from site “B.” So what does this mean for the results of a tariff?
If China stops buying soybeans from the US altogether, it means that unless they’re going to stop eating soybeans they will buy soybeans from Brazil. But if Brazil sells all of their soybeans to China, it means that Germany can no longer soybeans from Brazil. So where does Germany buy its soybeans from? Well, it seems that the US has beans that are not spoken for in this scenario…in other words, when we are talking about commodities a tariff mostly just reorganizes the list of who is buying from whom. If soybean prices are falling because China isn’t buying our soybeans, it means a great deal for Russia or Germany or whoever else is going to buy beans from us instead of from China’s new supplier. More than that, if global soybeans prices are falling because of tariffs then it means that everyone is getting cheaper soybeans because China is changing who they’re buying from. If that’s the case, then we really need to slap tariffs on everything and watch prices decline!
Let’s go back to elementary microeconomics. Adding a tariff is reflected in our product market supply and demand curves as a shift in the supply curve to the left: the quantity that producers are willing to supply at any price declines, because the price to the producer declines. Put a different way, the market price required to induce any particular quantity supplied rises by the amount of the tariff. Now, whether that causes market prices to rise a lot or a little, or quantity supplied to fall by a lot or a little, depends on the elasticities of supply and demand. If demand if fairly inelastic (which seems reasonable – you may be able to substitute for “beans” but it’s hard to substitute for “grains”), then you will see more of a price response than a quantity response, at least in the short run where the supply of beans is fairly inelastic. But that price response is up, not down.
By the way, this gets a little hard to illustrate with supply and demand curves, because with a tariff what you have are now two separate markets and separate prices for the same good. This is what confuses some people – if China is no longer buying from the US, doesn’t that mean that demand for US beans has declined, and therefore prices should decline? The crucial point is that we are talking here about commodity goods, and supplies are fairly interchangeable. If we are talking about Harley Davidson motorcycles, the answer is different because if Europe stops buying Harleys, they have to buy a different product altogether. In that case, the global price of “motorcycles” might be relatively unaffected, but the price of Harleys will rise (and the output decline) relative to other motorcycles. So, a tariff on Harley-Davidson motorcycles definitely hurts the US, but a tariff on soybeans – or even “US soybeans” since that is not a universal distinction – should have virtually no effect on US producers. And certainly, no effect on the global price of soybeans.
There are other reasons that grains prices may be declining. Since Brazil is a major producer of beans, the sharp decline in the Brazilian Real has pushed the US dollar price of beans lower (see chart, source Bloomberg). In the chart below, the currency is shown in Reals per dollar, and inverted. This is a much more important factor explaining the decline in grains prices, as well as one that could easily get worse before it gets better.
I think the discussion of the effects of tariffs has gotten a bit polluted since the decline in grains seems to coincide with the announcement of tariffs from China. I think the price decline here has fed that story, but it’s bad economics. Piecemeal tariffs on commodity products are not likely to appreciably change the supply and demand outcome, although it will result in rearranging the sources of product for different countries. Tariffs on non-commodity product, especially branded products with few close substitutes, can have much larger effects – although we ought to remember that from the consumer’s perspective (and in the measurement of consumer inflation), tariffs never lower prices faced by consumers although they can lower prices received by producers. This is why tariffs are bad – they cause higher prices and lower output, and the best case is no real change.
DISCLOSURE: Quantitative/systematic funds managed by Enduring Investments have both long and short positions in grains, and in particular long positions in Beans and Corn, this month.
Trade Surplus and Budget Deficit? Ouch.
The market gyrations of late are interesting, especially during the NCAA Basketball tourney. Normally, volatility declines when these games are on during the week, as traders watch their brackets as much as they do the market (I’ve seen quantitative analysis that says this isn’t actually true, but I’m skeptical since I’ve been there and I can promise you – the televisions on the trading floor are tuned to the NCAA, not the CNBC, on those days). Higher volatility not only implies that lower prices are appropriate in theory but it also tends to happen in practice: higher actual volatility tends to force leveraged traders to reduce position size because their calculation of “value at risk” or VAR generally uses trailing volatility; moreover, these days we also need to be cognizant of the small, but still relevant, risk-parity community which will tend to trim the relative allocation to equities when equity vol rises relative to other asset classes.
My guess is that the risk-parity guys probably respond as much to changes in implied volatility as to realized volatility, so some of that move has already happened (and it’s not terribly large). But the VAR effect is entirely a lagging effect, and it’s proportional to the change in volatility as well as to the length of time the volatility persists (since one day’s sharp move doesn’t change the realized volatility calculation very much). Moreover, it doesn’t need to be very large per trader in order to add up to a very large effect since there are many, many traders who use some form of VAR in their risk control.
Keep in mind that a sharp move higher, as the market had yesterday, has as much effect on VAR as a sharp move lower. The momentum guys care about direction, but the VAR effect is related to the absolute value of the daily change. So if you’re bullish, you want a slow and steady move higher, not a sharp move higher. Ideally, that slow and steady move occurs on good volume, too.
The underlying fundamentals, of course, haven’t changed much between Friday and Monday. The chance of a trade war didn’t decline – the probability of a trade war is now 1.0, since it has already happened. Unless you want to call an attack and counterattack a mere skirmish, rather than a trade war, there is no longer any debate about whether there will be conflict on trade; the only discussion is on magnitude. And on that point, nothing much has changed either: it was always going to be the case that the initial salvo would be stridently delivered and then negotiated backwards. I’m not sure why people are so delighted about the weekend’s developments, except for the fact that investors love stories, and the story “trade war is ended!” is a fun story to tell the gulli-bulls.
As a reminder, it isn’t necessary to get Smoot/Hawley 2.0 to get inflation. Perhaps you need Smoot/Hawley to get another Depression, but not to get inflation. The mere fact that globalization is arrested, rather than continuing to advance, is enough to change the tradeoff between growth and inflation adversely. And that has been in the cards since day 1 of the Trump Administration. A full-on trade war, implying decreased globalization, changes the growth/inflation tradeoff in a very negative way, implying much tighter money growth will be required to tamp down inflation, which implies higher interest rates. I’m not sure we aren’t still headed that way.
But there is a much bigger issue on trade, which also implies higher interest rates…perhaps substantially higher interest rates. We (and by ‘we’ I mean ‘he’) are trying to reduce the trade deficit while increasing the budget deficit sharply. This can only happen one way, and that is if domestic savings increases drastically. I wrote about this point first in 2010, and then re-blogged it in 2013, here. I think that column is worth re-reading. Here’s a snippet:
“And this leads to the worry – if the trade deficit explodes, then two other things are going to happen, although how much of each I can’t even guess: (I) protectionist sentiment is going to become very shrill, and fall on the ears of a President who is looking to burnish his populist creds, and (II) the dollar is going to be beaten like a red-headed stepchild (being a red-headed stepchild, I use that simile grudgingly).”
Well, it took a while to happen and I never dreamed the “President looking to burnish his populist creds” would be a (supposed) Republican…but that’s what we have.
Here’s the updated chart showing the relationship between these two variables.
It’s important to remember that you can’t have a trade account surplus and a financial account surplus. If someone sells a good to a US consumer, that seller holds dollars and they can either sell the dollars to someone else (in which case the problem just changes hands), buy a US good (in which case there’s no trade deficit), or buy a US security. If we need non-US persons to buy US securities, then we need to run a trade deficit. If we want to run flat on trade, then we either need to run a balanced budget or fund the difference out of domestic savings. A large increase in domestic savings implies a large decrease in domestic spending, especially if the Fed is now ‘dissaving’ by reducing its balance sheet. Inducing extra domestic savings also implies higher real interest rates. Now, this isn’t a cataclysmic result – more domestic savings implies more long-term domestic growth, although perhaps not if it’s being sopped up by the federal government – but it’s a very large shift to what the current balances are.
If you want to run a flat balance of trade, the best way to do it is to run a balanced federal budget. Going opposite directions in those two accounts implies uncomfortably large shifts in the account that makes up the difference: domestic savings, and large shifts in interest rates to induce that savings.
Are Rising Yields Actually a Good Thing?
I’ve recently been seeing a certain defense of equities that I think is interesting. It runs something like this:
The recent rise in interest rates, which helped cause the stock market swoon, is actually a good thing because interest rates are rising due to a strong economy and increasing demand for capital, which pushes up interest rates. Therefore, stocks should actually not mind the increase in interest rates because it’s an indication of a strong economy.
This is a seductive argument. It’s wrong, but it’s seductive. Not only wrong, in fact, but wrong in ways that really shouldn’t confuse any economist or strategist writing in the last twenty years.
Up until the late 1990s, we couldn’t really tell the main reason that nominal interest rates were rising or falling. For an increase in market rates there are two main potential causes: an increase in real interest rates, which can be good if that increase is being caused by an increasing demand for credit rather than by a decreasing supply, and an increase in inflation expectations, which is an unalloyed negative. But in 1995, we would have had to just guess which was causing the increase in interest rates.
But since 1997, we’ve had inflation-linked bonds, which trade on the basis of real yield. So we no longer have to guess why nominal rates are rising. We can simply look.
The chart below shows the decomposition of 10-year nominal yields since early December. The red line, which corresponds to the left scale, shows “breakevens,” or the simple difference between real yields and nominal yields; the blue line, on the right-hand scale, shows real yields. So if you combine the two lines at any point, you get nominal yields.
Real yields represent the actual supply and demand for the use of capital. That is, if I lend the government money for ten years, then in order to entice me to forego current consumption the government must promise that every year I will accumulate about 0.68% more ‘stuff.’ I can consume more in the future by not consuming as much now. To turn that into a nominal yield, I then have to add some premium to represent how much the dollars I will get back in the future, and which I will use to buy that ‘stuff’, will have declined in value. That of course is inflation expectations, and right now investors who lend to the government are using about 2.1% as their measure of the rate of deterioration of the value of the dollar.[1]
So, can we say from this chart that interest rates are mainly rising for “good” reasons? On the contrary! The increase in inflation expectations has been much steadier; only in the last month have real interest rates risen (and we don’t know, by the way, whether they’re even rising because of credit demand, rather than credit supply). Moreover – although you cannot see this from the chart, I can tell you based on proprietary Enduring Intellectual Properties research that at this level of yields, real yields are usually responsible for almost all of the increase or decrease in nominal yields.[2] So the fact that real yields are providing a little less than half of the selloff? That doesn’t support the pleasant notion of a ‘good’ bond selloff at all.
As I write this, we are approaching the equity market close. For most of the day, equities have been trading a bit above or a bit below around Monday’s closing level. While this beats the heck out of where they were trading overnight, it is a pretty feeble technical response. If you are bullish, you would like to see price reject that level as buyers flood in. But instead, there was pretty solid volume at this lower level. That is more a bearish sign than a bullish sign. However, given the large move on Friday and Monday it was unlikely that we would close near unchanged – so the last-hour move was either going to be significantly up or significantly down. Investors chose up, which is good news. But the bad news is that the end-of-day rally never took us above the bounce-high from yesterday’s last hour, and was on relatively weak volume…and I also notice that energy prices have not similarly rallied.
[1] In an article last week I explained why we tend to want to use inflation swaps rather than breakevens to measure inflation expectations, but in this case I want to have the two pieces add up to nominal Treasury yields so I am stuck with breakevens. As I noted in that article, the 2.1% understates what actual inflation expectations are for 10 years.
[2] TIPS traders would say “the yield beta between TIPS and nominals is about 1.0.”
The Mystery of Why There’s A Mystery
We have an interesting week ahead, at least for an inflation guy.
Of course, the CPI statistics (released this Friday) are always interesting but with all of the chatter about the “mystery” of inflation, it should draw more than the usual level of attention. That’s especially true since the mystery will cease to be a mystery fairly soon as even flawed indicators of inflation’s central tendency, such as the core CPI, turn back higher. This is not particularly good news for many pundits, who have declared the mystery to be solved with some explanation that implies inflation will stay low.
- “Amazon effect”
- Globalization
- “competition”
- Etc
The first of these I have addressed previously back in June (“The Internet Has Not Killed, and Will Not Kill, Inflation”). The second is a real effect, but it is a real effect whose effect peaked in the early 1990s and has been waning since then. I wrote something in our quarterly in Q4 last year, which is partly summarized here.
The “competition” objection is a weird one. It seems to posit that competition was pretty lame until recently, which is pretty strange. One argument along these lines is in this article by Steve Wunsch, who considers the increase in airline fees “stark evidence of a deflationary spiral in those ticket prices caused by antitrust-induced competition.” This is odd, since airlines were deregulated in 1978 and have in recent years become less competitive if anything with the mergers of Delta/Northwest in 2009, United/Continental in 2010, Southwest/AirTran in 2011, and US Airways/American Airlines in 2013. A flaccid antitrust response from the Justice Department has allowed quasi-monopolies to develop in some travel hubs, which has tended to push fares higher rather than lower. The chart below shows the relationship between Jet Fuel prices and the CPI for airfares (both seasonally adjusted) for the 20 years ended in 2014, along with the most-recent point from last month.
The highly-explanatory R-squared of 0.81 suggests that there is not much wiggle room in airline pricing. Airfares are, as you would expect under a competitive industry, roughly cost-plus with the main source of variance being jet fuel prices. This is true even though we would expect that spread to vary over time. As Mr. Wunsch would argue, the highly competitive nature of the industry is holding down the non-commodity price pressures in airfares.
The only problem is that if you extend this graph to include the last three years, the R-squared drops about 10 points:
In case it isn’t clear from that chart, the last three years have seen airfares increasingly above what we would expect from the level of jet fuel prices. The next chart makes that clear I hope by plotting the residual (and 12-month moving average to smooth out seasonal issues such as one that evidently happened last month) between the actual CPI-airfare and the level that would be predicted from the 1994-2014 relationship. As you can see, prices have been higher, and increasingly so, than we would have thought, until this last month or two – and I wouldn’t grab a lot of comfort from that yet.
Not only is this not “stark evidence of a deflationary spiral in those ticket prices caused by antitrust-induced competition,” it seems to be stark evidence of inflation in ticket prices caused by a reduction in competition thanks to airline mergers.
In reading these many articles, it always is somewhat striking to me: everybody thinks their answer is “the” answer to the mystery. But most of these authors really don’t sufficiently understand how inflation works, and what the data is showing. This is apparent to those who do understand these nuances, as an author might discuss (as the one mentioned above did) an “aberration” in cell phone inflation as if the experts are stupid for expecting inflation when cell phone services only go down. The author clearly misunderstands what the “aberration” referred to even is; in this case the aberration was an enormous one-month collapse in prices that had never been seen and has not been repeated since. (For those who are curious about the aberration, and why it occurred, and why it is likely a methodology issue rather than sign of spiraling deflation in wireless services you can see my discussion of it here.)
The mystery is simple – the Fed’s models don’t work, and don’t take into account the fact that lower interest rates cause lower money velocity. They rely on a Phillips Curve effect that they think is broken because they don’t understand that the Phillips Curve relates wages and unemployment, not consumer prices and unemployment. They focus on a flawed measure like PCE rather than on something like Median CPI which, coincidentally, is a lot higher and suggests more price pressures. The mystery isn’t why inflation isn’t rising yet – the mystery is why they think there’s a mystery.
Some Further (Minor) Thoughts on the Phillips Curve
Before I begin, let me say that if you haven’t read yesterday’s article, please do because it represents the important argument: the Phillips Curve doesn’t need rehabilitating, because it is working fine. In fact, I would argue that the Phillips Curve – relating wages to unemployment – is a remarkably accurate economic model prediction. The key chart from that article I reproduce here, but the article (which is brief) is worth reading.
Following my publication of that article, I had a few more thoughts that are worth discussing on this topic.
The first is historical. It’s incredibly frustrating to read article after article incorrectly stating what the Phillips Curve is supposed to relate. Of course one writer learns from another writer until what is incorrect becomes ‘common knowledge.’ I was fortunate in that, 30 years ago, I had excellent Economics professors at Trinity University in San Antonio, and I was reflecting on that fact when I said to myself “I wonder if Samuelson had it right?”
So I dug out my copy of Economics by Samuelson and Nordhaus (the best-selling textbook of all time, I believe, and the de rigeur Intro to Economics textbook for generations of economists). My copy is the 12th Edition, so perhaps they have corrected this since then…but on page 247, there it is – the Phillips Curve illustrated as a “tradeoff between inflation and unemployment.” Maybe that is where this error really propagated – with a Nobel Prize-winning economist making an error in his incredibly widely-read text! Interestingly, the authors don’t reference the original Phillips work, but refer to “writers in the 1960s” who made that connection, so to be fair to Samuelson and Nordhaus they were possibly already repeating an error that had been made even earlier.
My second point is artistic. In yesterday’s article, I said “The Phillips Curve…simply says that when labor is in short supply, its price goes up. In other words: labor, like everything else, is traded in the context of supply and demand,…” But students of economics will note that the Phillips Curve seems to obfuscate this relationship, because it is sloping the wrong way for a supply curve – which should slope up and to the right rather than down and to the right. This can be remedied by expressing the x-axis of the Phillips Curve differently – making it the quantity of labor demanded rather than the quantity of labor not demanded…which is what the unemployment rate is. So the plot of wage inflation as a function of the Employment Rate (as opposed to the Unemployment Rate) has the expected shape of a supply curve. More labor is supplied when the prices rise.
Again, this is nuance and not a really important point unless you want your economics to be pretty.
My third point, though, is important. One member of the bow-tied fraternity of Ph.D. economists told me through a friend that “the Phillips Curve has evolved to the relationship between Unemployment and general prices, not simply wages.” I am skeptical of any “evolution” that causes the offspring to be worse-adapted to the environment, but moreover I would argue that whoever led this “evolution” (and as I said above, it looks like it happened in the 1960s) didn’t really understand the way the economy (and in particular, business) works.
There is every reason to think that wages should be tied to available labor supply because one is the price of the other. That’s Microeconomics 101. But if unemployment is going to be a good indicator of generalized price inflation too, then it means that prices in the economy are essentially set as the price of the labor input plus a spread for profit. That is not at all how prices are set. Picture the businessperson deciding how to set prices. According to the “evolved Phillips Curve” understanding, this business owner looks at the wages he/she is paying and then sets the price of the product. But that’s crazy. A business owner considers labor as one input, as well as all of the other inputs, improvements in productivity in producing this good or service in question, competitive pressures, and the general state of the national and local economy. It would be incredible if all of these factors canceled out except for wage inflation, wouldn’t it? So in short, while I would expect that unemployment might have some explanatory power for inflation, I wouldn’t expect that explanatory power to be very strong. And, in fact, it isn’t. (But this isn’t new – it never has had any power.)
.
Come see our new store at https://store.enduringip.com!
The Phillips Curve is Working Just Fine, Thanks
I must say that it is discouraging how often I have to write about the Phillips Curve.
The Phillips Curve is a very simple idea and a very powerful model. It simply says that when labor is in short supply, its price goes up. In other words: labor, like everything else, is traded in the context of supply and demand, and the price is sensitive to the balance of supply and demand.
Somewhere along the line, people decided that what Phillips really meant was that low unemployment caused consumer price inflation. It turns out that doesn’t really work (see chart, source BLS, showing unemployment versus CPI since 1997).
Accordingly, since the Phillips Curve is “broken,” lots of work has been done to resurrect it by “augmenting” it with expectations. This also does not work, although if you add enough variables to any model you will eventually get a decent fit.
And so here we are, with Federal Reserve officials and blue-chip economists alike bemoaning that the Fed has “only one model, and it’s broken,” when it never really worked in the first place. (Incidentally, the monetary model that relates money and velocity (via interest rates) to the price level works quite well, but apparently they haven’t gotten around to rediscovering monetarism at the Fed).
But the problem is not in our stars, but in ourselves. There is nothing wrong with the Phillips Curve. The title of William Phillips’ original paper is “The Relation between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957.” Note that there is nothing in that title about consumer inflation! Here is the actual Phillips Curve in the US over the last 20 years, relating the Unemployment Rate to wages 9 months later.
The trendline here is a simple power function and actually resembles the shape of Phillips’ original curve. The R-squared of 0.91, I think, sufficiently rehabilitates Phillips. Don’t you?
I haven’t done anything tricky here. The Atlanta Fed Wage Growth Tracker is a relevant measure of wages which tracks the change in the wages of continuously-employed persons, and so avoids composition effects such as the fact that when unemployment drops, lower-quality workers (who earn lower wages) are the last to be hired. The 9-month lag is a reasonable response time for employers to respond to labor conditions when they are changing rapidly such as in 2009…but even with no lag, the R-squared is still 0.73 or so, despite the rapid changes in the Unemployment Rate in 2008-09.
So let Phillips rest in peace with his considerable contribution in place. Blame the lack of inflation on someone else.
.
Come see our new store at https://store.enduringip.com!
Profits and Health Care: A Beneficial Connection
I usually try to avoid political commentary in this space, because it has become so personal to so many people. If I point out that a particular program of the “left” is smart, or cleverly put together, then half of my readership is annoyed; if I point out the same about the right, then half of my readership is angry. It doesn’t really make sense to waste article space except on those occasions when a policy has a clear effect on inflation over time, such as when the structure of the ACA made it clear that it would put upward pressure on inflation (as I pointed out in 2013) or in response to someone else’s flawed analysis of a policy, as I did last year when I tackled the San Francisco Fed for their weak argument about how the ACA would hold down inflation because the government would demand lower prices. Actually, there is no policy I have written about more than the ACA over the years – but again, this was economic commentary and not political commentary.
This article will be short, but different in that I am writing it to express frustration with the absolute lack of intellectual clarity on the part of the Republicans in making a particular argument that immediately impacts the debate over health care but also extends far into other policies. And, because the argument is simple, direct, and has tremendous empirical support, I couldn’t restrain myself. I expect this article will not be picked up and syndicated in its usual channels since it isn’t directly about economics or markets, but it needed to be said.
I’ve been stewing about this topic since Tuesday (March 14th), when I happened to catch part of the daily White House press briefing. Press Secretary Sean Spicer was asked a question about the President’s health care proposal, and tap danced away from the question:
Q Thanks, Sean. You mentioned the call with the CEO of Anthem Health. Can you tell me what this proposal of the President means for health insurance companies? Will their profits go up or down under the President’s proposal?
SPICER: Well, I don’t think that’s been the focus of the President’s proposal. It’s not about them, it’s about patients. But I think what it means for them is that they finally get to create more choice and more plans and allow people to choose a plan that fits them. Right now, they don’t have that choice. And, frankly, in more and more markets, companies like Anthem, UnitedHealth, Signa are pulling out — Aetna — because they don’t have the choice and because of the government mandate. I think what we want to do is allow competition and choice to exist so that they can offer more options for the American people.
Q But will those companies make more money under the President’s plan or less?
SPICER: I don’t know the answer to that. That’s not been the focus of what we’re doing now. And at the end of the day, right now they’re pulling out of market after market, leaving the American people with fewer and fewer choices. So right now it’s not a question of — from the last I checked, I think many of them were doing pretty well, but it’s the American people and its patients that are losing under the current system. So I think that there’s a way you can do a little of both.
Spicer’s response was the usual drivel that the Republicans have adopted when they run in fear from any question that includes the word “profits.” To summarize, the question was basically, “you’re doing this to throw a sop to fat-cat insurance companies, aren’t you?” and the answer was “we don’t think about that. No idea. Profits? Who said anything about profits? It’s about patients and choice. And, if anyone gets more profits, it wasn’t on purpose and we didn’t have anything to do with it.”
But this was actually a softball question, and the answer ought to have been something like this:
Q But will those companies make more money under the President’s plan or less?
BIZARRO SPICER: Well, I hope so. After all, the insurance companies want every person in America to have health care – which is the same thing that we want – because the more people they sell their product to, the more money they can make. The insurance companies want to sell insurance to every person in the U.S. The insurance companies also want costs to be lower, and constantly strive to lower the cost of care, because the lower that costs are, the more profit they can make in the short run. But they don’t want lower costs at the expense of health – clearly, the best outcome for their profits is that most people covered by insurance are healthy and so don’t require the insurance they’ve paid for. So, if we just get out of the way and let companies strive for better profits, we are likely to get more coverage, lower costs, and a healthier population, and that is the goal of the President’s plan.
The reason we don’t already have these things is that laws we have previously passed don’t allow insurance companies to offer certain plans, to certain people, which both sides want but which politicians think are “unfair” for one reason or another. Trying to create a certain preconceived Utopian outcome while limiting profits of insurance companies is what caused this mess in the first place.
If you want to beautify gardens in this city, does it make sense to limit the amount of money that gardeners can make? If you did, you would find fewer gardens got tended, and gardeners would not strive to make improvements that they didn’t get paid for. We can see this clearly with gardeners. Why is it so hard to understand with the companies that tend to the nation’s health? Next question.
For some reason, Republicans think that saying “profits are good” is the same thing as saying “greed is good” and leads to caricatures of conservatives as cigar-smoking industrialists. But while at some level it is the desire for a better material outcome – which I suppose is greed, but aren’t there degrees of greed? – that drives the desire for profit, we cannot dismiss the power of self-interest as a motive force that has the effect of improving societal outcomes. “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest,” after all.
Of course, Republicans must also remember that profit without competition is a different animal. If an insurance company creates an innovation that lowers medical care costs, but does not face competitive pressure, then the benefit of the innovation accrues to the company alone. There is no pressure in such circumstances for the company to lower the price to the customer. But consider what happened to air fares after the deregulation of 1978, or to the cost of telephone service when the AT&T monopoly was broken up in 1984, as competition was allowed and even encouraged. Competition, and the more brutal the better, is what causes companies to strive for an edge through innovation, and it’s also what causes the benefit of that edge to eventually be accrued by the end customer. The government didn’t invent cell phones. Motorola did, in order to try and gain an edge against AT&T,[1] but until the telephone monopoly was broken up there were no commercial versions of the cell phone. The first cell phones cost $10,000 in 1983, about $25,000 in today’s dollars, but now they are ubiquitous and cost about 2% as much in real terms. But this didn’t happen because of a government program to drive down the cost of cell phones. It was the profit motive, combined with competition. All that government did was create the conditions that allowed innovation and competition to happen. And wouldn’t we like health care to be as ubiquitous and cheap as cell phones are?
This is not a hard thing to get right. It isn’t hard for people to understand. But for some reason, it seems incredibly hard for politicians to believe.
Note that nothing I have written here should be construed as an opinion about the President’s health care plan, which I have not read. My remarks are only meant to reflect on the utter inability of Republicans to properly convey the reasons that a different approach – one where the government’s involvement is lessened, rather than increased – would make more sense.
[1] The first cell phone call was made by the inventor, Martin Cooper at Motorola, who called his competition with it: the head of the cellular program at AT&T. According to him, he said “Joel, I’m calling you from a cellular phone, a real cellular phone, a handheld, portable, real cellular phone” and he said it got really quiet on the other end of the line.
Good Models and Bad Models
I have recently begun to spend a fair amount of time explaining the difference between a “good model” and a “bad model;” it seemed to me that this was a reasonable topic to put on the blog.
The difference between a good model and a bad model isn’t as obvious as it seems. Many people think that a “good model” is one that makes correct predictions, and a “bad model” is one that makes bad predictions. But that is not the case, and understanding why it isn’t the case is important for economists and econometricians. Frankly, I suspect that many economists can’t articulate the difference between a good model and a bad model…and that’s why we have so many bad models floating around.
The definition is simple. A good model is one which makes good predictions if high-quality inputs are given to the model; a bad model is one in which even the correct inputs doesn’t result in good predictions. At the limit, a model that produces predictions that are insensitive to the quality of the inputs – that is, whose predictions are just as accurate no matter what the inputs are – is pure superstition.
For example, a model of the weather that depends on casting chicken bones and rat entrails is a pretty bad model since the arrangement of such articles is not likely to bear upon the likelihood of rain. On the other hand, a model used to forecast the price of oil in five years as a function of the supply and demand of oil in five years is probably an excellent model, even though it isn’t likely to be accurate because those are difficult inputs to know. One feature of a good model, then, is that the forecaster’s attention should shift to the forecasting of the inputs.
This distinction is relevant to the current state of practical economics because of the enormous difference in the quality of “Keynesian” models (such as the expectations-augmented Phillips curve approach) and of monetarist models. The simplest such monetarist model is shown below. It relates the GDP-adjusted quantity of money to the level of prices.
This chart does not incorporate changes in money velocity (which show up as deviations between the two lines), and yet you can see the quality of the model: if you had known in 1948 the size of the economy in 2008, and the quantity of M2 money there would be in 2008, then you would have had a very accurate prediction of the cumulative rate of inflation over that 60-year period. We can improve further on this model by noting that velocity is not random, but rather is causally related to interest rates. And so we can state the following: if we had known in 2007 that the Fed was going to vastly expand its balance sheet, causing money supply to grow at nearly a 10% rate y/y in mid-2009, but at the same time 5-year interest rates would be forced from 5% to 1.2% in late 2010, then we would have forecast inflation to decline sharply over that period. The chart below shows a forecast of the GDP deflator, based on a simple model of money velocity that was calibrated on 1977-1997 (so that this is all out-of-sample).
That’s a good model. Now, even solid monetarists didn’t forecast that inflation would fall as far as it did – but that’s not a failure of the model but a failure of imagination. In 2007, no one suspected that 5-year interest rates would be scraping 1% before long!
Contrariwise, the E-A-Phillips Curve model has a truly disastrous forecasting history. I wrote an article in 2012 in which I highlighted Goldman Sachs’ massive miss from such a model, and their attempts to resuscitate it. In that article, I quoted these ivory tower economists as saying:
“Economic principles suggest that core inflation is driven by two main factors. First, actual inflation depends on inflation expectations, which might have both a forward-looking and a backward-looking component. Second, inflation depends on the extent of slack (or spare capacity) in the economy. This is most intuitive in the labor market: high unemployment means that many workers are looking for jobs, which in turn tends to weigh on wages and prices. This relationship between inflation, expectations of inflation and slack is called the “Phillips curve.”
You may recognize these two “main factors” as being the two that were thoroughly debunked by the five economists earlier this month, but the article I wrote is worth re-reading because it describes how the economists re-calibrated. Note that the economists were not changing the model inputs, or saying that the forecasted inputs were wrong. The problem was that even with the right inputs, they got the wrong output…and that meant in their minds that the model should be recalibrated.
But that’s the wrong conclusion. It isn’t that a good model gave bad projections; in this case the model is a bad model. Even having the actual data – knowing that the economy had massive slack and there had been sharp declines in inflation expectations – the model completely missed the upturn in inflation that actually happened because that outcome was inconsistent with the model.
It is probably unfair of me to continue to beat on this topic, because the question has been settled. However, I suspect that many economists will continue to resist the conclusion, and will continue to rely on bad, and indeed discredited, models. And that takes the “bad model” issue one step deeper. If the production of bad predictions even given good inputs means the model is bad, then perhaps relying on bad models when better ones are available means the economist is bad?
August, Productivity, and Prices
I really don’t like August. It’s nothing about the weather, or the fact that the kids are really ready to be back in school (but aren’t). I just really can’t stand the monkey business. August is, after December, probably the month in which liquidity is the thinnest; in a world with thousands of hedge funds this means that if there is any new information the market tends to have dramatic swings. More to the point, it means that if there is not any new information, the speculators make their own swings. A case in point today was the massive 5% rally in energy futures from their lows of the day back to the recent highs. There was no news of note – the IEA said that demand will balance the oil market later this year, but they have said that in each of the last couple of months too. And the move was linear, as if there had been news.
Don’t get me wrong, I don’t care if traders monkey around with prices in the short run. They can’t change the underlying supply and demand imbalance and so it’s just noise trading for noise trading’s sake. What bothers me is that I have to take time out of my day to go and try to find out whether there is news that I should know. And that’s annoying.
But my whining is not the main reason for this column today. I am overdue to write about some of the inflation-related developments that bear comment. I’ll address one of them today. (Next week, I will probably tackle another – but Tuesday is also CPI day, so I’ll post my usual tweet summary. Incidentally, I’m scheduled to be on What Did You Miss? on Bloomberg TV at 4pm ET on Tuesday – check your local listings).
I don’t spend a lot of time worrying about productivity (other than my own, and that of my employees). We are so bad at measuring productivity that the official data are revised for many years after their release. For example, the “productivity miracle” of the late 1990s, which drove the Internet bubble and the equity boom into the end of the century, was eventually revised away almost completely. It never happened.
The problem that a lot of people have with thinking about productivity is that they confuse the level of productivity with its pace of increase. So someone will say “of course the Internet changed everything and we got more productive,” when the real question is whether the pace of productivity increase accelerated. We are always getting more productive over time. There are always new innovations. What we need to know is whether those innovations and cost savings are happening more quickly than they used to, or more slowly. And, since the national accounts are exquisitely bad at picking up new forms of economic activity, and at measuring things like intellectual property development, it is always almost impossible to reject in real time the hypothesis that “nothing is changing about the rate of productivity growth.” Therefore, I don’t spend much time worrying about it.
But, that being said, we should realize that if there is a change in the rate of productivity growth it has implications for growth, but also for inflation. And recent productivity numbers, combined with the a priori predictions in some quarters that the global economy is entering a slow-productivity phase, have started to draw attention.
Most of that attention is focused on the fact that poor productivity growth lowers overall real output. The mechanism there is straightforward: productivity growth plus population growth equals real economic output growth. (Technically, more than just population growth it is working-age population growth times labor force participation, but the point is that it’s an increase in the number of workers, compounded by the increase in each worker’s productivity, that increases real output). Especially if a populist backlash in the US against immigration causes labor force growth to slow, a slower rate of productivity growth would compound the problem of how to grow real economic growth at anything like the rate necessary to support equity markets or, for that matter, the national debt.
But there hasn’t been as much focus on the other problem of low productivity growth, if indeed we are entering into that sort of era. The other problem is that low productivity growth causes higher prices, all else equal. That mechanism is also straightforward. We know that money growth plus the change in money velocity equals real output growth plus an increase in prices: that is, MV≡PQ. If velocity is mean-reverting, then the decline in real growth precipitated by a decline in labor productivity, in the context of an unchanged rate of increase in the money supply, implies higher prices. That is, if ΔM is constant and ΔV is zero and ΔQ declines, then ΔP must increase.
One partial offset to this is the fact that a permanent decline in productivity growth rates would lower the equilibrium real interest rate, which would lower the equilibrium money velocity. But that is a one-time shift while the change in trend output would be lasting.
In fact, it wouldn’t be unreasonable to suppose that the change in interest rates we have seen in the last few years is mostly cyclical but may also be partly secular. This would imply a lower equilibrium level of interest rates (although I don’t mean to imply that anything is near equilibrium these days), and a lower equilibrium level of monetary velocity. But there are a lot of “ifs” in that statement.
The biggest “if” of all, of course, is whether there really is a permanent or semi-permanent down-shift in long-term productivity growth. I don’t have a strong opinion on that, although I suspect it’s more likely true that the current angst over low productivity growth rates is just the flip side of the 1990s ebullience about productivity. We’ll know for sure…in about a decade.
