I have been thinking about an issue that puzzles me – it is about inflation in an Austrian School style bust.
Here is the story. If we think about a stylized Austrian school boom-bust then the story more or less is that easy money leads to an unsustainable boom that eventually – for some reason – will lead to a bust.
What people often fail to realize is that the Austrian business cycle theory basically is a supply side story.
Austrians will hate it, but you can tell much of the story within an AS/AD framework. The graph below is an illustration of this.
What happens is that the central bank cuts interest rates below the Wicksellian natural interest rate. Investors are tricked into thinking that it is the natural interest rates that has fallen and as a result investments are increased. Austrians will of course object by saying “it is not a overinvestment theory, but a malinvestment theory”. Yes, that is right, but that is not relevant for the question I want to look at here.
The boom happens not because of higher demand, but because of over (and mal) investment. The production capacity of the economy is hence expanded – the AS curve shifts to the right during the Austrian boom and production increases from Y1 to Y2. We ignore the demand effects – so we keep the AD unchanged – as the Austrians really are not paying much attention to this part of the story anyway (and yes, I am aware the there is relative demand story – private consumption vs investments).
Notice what happens with the price level initially. Prices drop from P1 to P2. Obviously that would not necessarily have to be the case if the AD curve also have shifted to the right as well (but that is not important for the story here). However, this pretty well illustrates the Austrian story that “headline” inflation will not necessarily increase during the boom. What happens – and we can obviously not realize that by just looking at AD and AS curves – is that we get what Austrians call relative inflation. Some prices rise, but the aggregate price level does not necessarily increase.
So far so good. I know Austrian economists would say that I told the story in the “wrong way”, but I guess they will agree on the main points – Austrian Business Cycle Theory is mostly about the supply side of the economy and that the aggregate price level will not necessarily have to increase during the boom phase.
Now we turn to the bust phase…
At some point investors realise that they have made a mistake – the natural interest rate has not really dropped. Therefore, what they thought were good and profitable investments are not really that great. So as a result investors cut back investments – after the “bubble” have bursted a large part of the production capacity in the economy is worthless. This is a negative supply shock! The AS shift back leftwards.
What is the result of this? Well, it is simple – the price level increases from P2 to P1. We get higher inflation. This might seem counterintuitive to most people – that the bust leads to higher rather lower inflation – but remember this is due fact that the Austrian boom-bust cycle primarily is a supply side story.
‘Benign’ inflation should be welcomed
And this brings me to what I really wanted to say. An increase inflation should be welcomed if it reflects a rational and undistorted reaction to investors realising that they have made a mistake. That is exactly what happens in an Austrian style bust. We might get relative deflation/disinflation, but the aggregate price level increases due to the negative supply shock.
Therefore, when Austrians often argue that the bust should be allowed to play out without any interference from the government or the central bank then that logically mean that they should welcome an increase in inflation in the bust phase of the business cycle. That obvious is not that same saying that monetary policy should be eased in the bust phase, but inflation should nonetheless be allowed to increase as we get “benign” inflation.
However, in my view that would mean that it would be wrong from an Austrian perspective for the central bank to tighten monetary policy in response to rise in (supply) inflation during the bust. Those Austrian economists who favour NGDP level targeting – like Anthony Evans and Steve Horwitz – would likely agree, but what about the “internet Austrian”? And what about Bob Murphy or Joe Salerno?
Obviously the story I have told above is a caricature of the Austrian Business Cycle theory, but I think there is a relevant discussion here that need to be addressed. Is the aggregate price level likely to rise in the bust phase as natural consequence of market forces being allowed to run it cause?
The reason that I think this debate is important is that some Austrians spend a lot of time arguing that the deflationary tendencies that we see for example in Europe at the moment are a natural and necessary bursting and deflating of a bubble. However, IF we indeed were in the bust phase of a Austrian style business cycle then we would not be seeing deflationary tendencies. We would in fact be seeing the opposite – we would see HIGHER inflation, but at the same time relative deflation.
Obviously this is not what we are seeing in the US and Europe today – inflation in both the US and the euro zone is well-below what it was during the “boom years”. That mean that we are not in the bust phase of an Austrian style boom-bust. There might very well have been a boom-bust initially (I believe that was the case in some European countries for example), but we have long ago moved to another phase – and that is what Hayek termed secondary deflation – a downturn in the economy caused by an monetary contraction.
PS Take a look at what happened in the US in 2007-8. Overall inflation did in fact increase as the economy was slowing, while we at the same time had relative deflation in the form of falling property prices. However, starting in the Autumn of 2008 we clearly saw across the board deflationary tendencies – here it is pretty clear that we entered a secondary deflationary phase caused by a monetary contraction. This is consistent with an Austrian interpretation of the Great Recession, but it is not a story I have heard many (any??) Austrians tell. And of course it is not necessarily the story I would tell – even though I think there is a lot of truth in it.
PPS The graphs above could indicate that both production and prices shifts back to the initial starting point during the bust. That obviously would not have to be the case as I here have ignored the shift in the AD curve and as any Austrian would note the AS/AD framework is not telling us anything about relative prices.
Greg Ransom
/ April 7, 2013It _is_ in part an aggregate demand story — part of the story is that both consumption and investment expand beyond what possibly will be produced across time, and both production and consumption will contract, with consumption moving toward outputs with a shorter time horizon.
Lars Christensen
/ April 7, 2013Thanks for the comment Greg. I hoped you would comment on my post given your huge insight to the Hayekian story.
I agree on the the demand side of the story. However, there is certainly a supply side story as well and that should lead to higher inflation in the bust phase. Would you agree on that? Would you be able to point my to Hayek commenting on this issue?
Steve Horwitz
/ April 7, 2013I think the problem here Lars is that you start by talking in terms of “over” not “mal” investment. Portraying the Austrian cycle as an expansion of aggregate supply (and I think AS/AD can be useful for illustrating some basic macro stuff) requires that you think of total investment as having expanded to drive that AS curve to the right. However, when you think in terms of a structure of production argument, what’s happening is that we are diverting investment expenditures from stages of production closer to consumption to those stages more remote from consumption. The total amount of investment need not change and the AS curve need not shift.
And as Roger Garrison points out, the false interest rate signal also discourages saving and encourages consumption by consumers. The Hayekian triangle is pulled at both ends, with more resources to earlier stages and more consumption expenditures, both of which come from the middle stages (and from household savings). For a short period, he argues, we exceed the PPF (if that’s understood as a *sustainable* level of production) as the increased spending at both ends is undertaken AS IF the PPF had expanded. That might well look more like a shift in AD than in AS.
So whatever the benefits of AS/AD for capturing some basic stuff, I think it is cannot capture what’s happening in the world of the Hayekian triangle and Austrian capital structure, which in turn suggests that your story here might not be consistent with the theory.
troopa
/ April 7, 2013I always thought that AS is actually staying more or less the same. during boom – just changing “composition”. Mr. Horowitz, how do you see the “demand” part”? Somehow, I always thought that without some sort of “overshooting” – it wouldn”t be called a “boom”?
Justin Merrill
/ April 7, 2013Lars, I think you have some good points, but I also want to clarify some things too. I will refrain from being pedantic or criticizing your version of the ABCT; but I think it is worth mentioning the focus on relative prices and the role of money/bank credit in the cycle. The price of inputs will increase relative to outputs in the boom phase, even if the AS increases. Also, since it is caused by an increase in money, and we know that money increases affect the price level and NGDP, it seems hard to argue that the price level and the demand curve would remain unaffected by the monetary affects.
I also don’t know if I’d consider resources idled by the business cycle a supply shock.
But, that being said, I think you are correct in regards to the secondary deflation. I think you are a little off on the benign inflation part, or at least in its phase of the business cycle. I think the price inflation shows up not in the recession, but just before it. The investment that exceeded savings causes inputs like energy and commodities to spike and/or interest rates to rise. This is what makes the projects unprofitable. The problem is compounded by central banks that are inflation targeting tightening when they shouldn’t be!
This I think is maybe the point of your article, which is where I strongly agree with you. I think the Fed was too tight in ’07-’08 as well as 1928-1930. The question I want to ask Rothbardians is, “When do you think the Federal Reserve has been too tight?”
Central Banks do tend to err on the side of inflation, but when they get it wrong in either direction it really hurts.
Nicolas Cachanosky
/ April 7, 2013I agree with Steve. I would say the main problem pointed by Hayek is “how” goods are being produced, not “how much” of them are being produced.
If I were to put ABCT in terms of AS/AD, I would think of structural changes “inside” AS (more than movements of AS) and eventually movements of AD. Of course, the problem is that structural changes “inside AS” cannot be shown in the AS/AD graph and this is where this tool falls short in this case.
Jon Finegold
/ April 7, 2013Lars,
This may interest you: Hayek–Mises Theory as a Demand-Side One.
Lars Christensen
/ April 8, 2013Steve, Nicolas, Justin and Jon. You are all excellent economists and well-schooled in Austrian thinking. I am aware that I am not telling the right story about relative prices and the capital structure etc. However, I think I have a point.
Lets forget about the boom and lets focus on the bust. In the bust in an ABCT model we basically have a natural disaster – part of the production capacity is blown up. Isn’t it so? And if that is correct then that is a negative supply shock. Hence, prices must increase in the bust.
Anyway, lets have a vote – will prices in general increase or decrease during the bust in ABCT?
Nicolas Cachanosky
/ April 8, 2013Lars,
ok…. I see what you are trying to deal with.
The bust, however, can also see a collapse in money supply. That, plus stock sales to obtain liquidity, pushes prices down. If factors of production (ie labor) see their income dwindle because the project is not as profitable as was expected, if these resources are “fired”, then demand also falls. The supply problems have an effect on the demand side. If the value of supply falls, then how much can be demanded is affected as well.
Nicolas Cachanosky
/ April 8, 2013By a “collapse” in money supply I mean the outcome of firms being unable to service their debts to the financial sector.
Mike Sax
/ April 8, 2013Lars a very interesting, and, for me at least, timely piece as I’ve recently been trying to wrap my head around ABCT.
I have thought the same thing as you: that ABCT is a supply side story-clearly it’s real in that the productive capacities of the economy are damaged short term.
I actually just came across this notion of “relative” inflation or deflation but you actually provided me some clarity on it.
I just wrote a piece where I contrasted ABCT with Scott’s recent monetary primer posts.
http://diaryofarepublicanhater.blogspot.com/2013/04/what-causes-business-cycle-sumner-vs.htm
It seems to me that there’s some convergence between what I will take some liberty in calling “left wing economics” vs. “right wing economics.”
Both see the causes of the business cycle in more real than nominal terms. Certainly the prescriptions of ABCT couldn’t be more different than the Post Keynesains, some of the descrpition of how it happens is similar.
I know that you and most of the Austrians too for certain, don’t have a high view of PKers, but that’s not the point I’m getting at. Rather in explaining the business cycle both see real vs. nominal causes of recessions.
http://diaryofarepublicanhater.blogspot.com/2013/04/some-thoughts-on-krugman-vs-stockman.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+DiaryOfARepublicanHater+%28Diary+of+a+Republican+Hater%29
I’d also note that some major PKers actually started out as Austrains-Abba Lerner, et. al. Again my point isn’t to yet againlitigate the battle between these schools.
Mike Sax
/ April 8, 2013Your piece inspired this:
http://diaryofarepublicanhater.blogspot.com/2013/04/lars-christensen-austrianism-as-supply.html
Mike Sax
/ April 8, 2013When I speak of the “Left” and “Right” in economics converging I mean doing so over and against the “Center.” I think both Krugman and Sumner could reasonbly be placed closer to the Center. Both see the problems of the business cycle as more nominal than real.
anthonyjevans
/ April 8, 2013If you did a Venn diagram of economists that utilise AD-AS curves and those who buy into ABC I doubt there’s many in the middle. But I like the attempt to bridge them.
I’m surprised that you’re not using the Cowen/Tabarrok version because that seems to be more friendly to a market monetarist view. It also makes it more clear that AD shifts play a part in the Austrian story. I think there’s a danger that because some Austrian economists deny the conceptual validity of “Aggregate Demand” that they imply that shifts in the AS curve tell the whole story. I think this is a straw man though.
Another point about the Cowen/Tabarrok version is that the SRAS curve draws in money illusion. This reinforces the idea that for the boom phase at least it’s possible to view it as a movement along an AS curve rather than a shift in it (I think your analysis would be strengthened with a clear distinction between SRAS and the underlying potential growth rate).
For what it’s worth I find the Cowen/Tabarrok version a useful way to see recent events, but I’d point out that the stylised facts aren’t necessarily deflationary – in the UK at least CPI has been consistently well above target. As you say, if this is a result of a negative supply shock then it isn’t necessarily an indication of loose monetary policy, and if you want to target NGDP then it’s not grounds to tighten.
For me the crux of the issue is that ABC implies that at the height of the boom there’s a spike in consumer prices. This “moment” is actually the bust phase of the cycle. So yes, in the bust phase prices can rise. And indeed there is then a potential for a secondary recession (or “cumulative rot”). What isn’t well understood (at least by me) is what comes in between…
To answer your question, it depends on (at least) three things: (1) how long does the “spike” in consumer prices last? (2) is the secondary recession a permissible part of the ABC story (3) what’s the difference between the “bust” phase and the “recovery” phase of the cycle?
If the “bust” is the inevitable consequence of the boom then it’s not determined by policy. Therefore if the secondary recession is cased by bad policy (e.g. allowing NGDP expectations to contract) then it’s not part of the bust phase. I’d call it part of the “recovery” phase (albeit not a good recovery).
I concur that MV=PY is the best lens through which to view it. When I use AD-AS in class in relation to recent events, I show a negative AD shock *and* a negative shock to potential real GDP. Hence the present situation of below trend (but not necessarily below potential) growth, and above target inflation. I think that it was a mistake to allow NGDP to fall, but having made that mistake there’s only limited (and not necessarily desirable) gains in trying to increase it back to a previously unsustainable level.
dajeeps
/ April 8, 2013Thanks for posting this. I have an issue in general with, and maybe it’s a label problem, ideas that say things like people just have no way of understanding what they are doing if short term nominal rates differ in the short run. It seems contrary to free market philosophy that people would all of a sudden decide to buy a Ferrari instead of start a hardware business or put a kid through college, for example; and it leads, I think, to all kinds of misunderstanding about controlling the investment behavior of individuals instead of keeping a neutral monetary policy.
Becky Hargrove
/ April 9, 2013Lars,
The reason I appreciate this post so much is that the so called lessons of malinvestment or overinvestment haven’t really been learned in the U.S. Too many people would rather wait till some “lose” the game (bankers and borrowers) and then start all over again with the same board! First lesson: stabilize the board with an NGDP thought process . Second lesson: for God’s sake, start serving the portion of the market that couldn’t afford the original Luddite definition of (construction/building) in the first place, instead of blaming the “losers” and the governments/central banks that tie themselves itself into knots trying to stabilize what people insisted they wanted in the first place.
Bob Murphy
/ April 13, 2013Try this article Lars.
Majror_Freedom
/ April 13, 2013Lars, ABCT does not predict that price inflation will always be higher during busts than it is during booms. During busts, Austrian theory explicitly recognizes that there are many competing and exacerbating factors that could make price inflation go up or down: For example the central bank’s response to recession, people’s new cash preference, and so on.
Using AS/AD analysis to paint ABCT as a pure supply side only theory, is a flawed approach. Now I know that you said that Austrians would say this is the “wrong way” to look at it, but that doesn’t mean your correct prediction on this score means you’re justified in your approach.
ABCT is not purely supply side, such that it holds the money supply and volume of spending constant throughout, such that it predicts higher prices during negative shocks to AS. ABCT deals with the structure of production, period. It does not make any statements with regards to the directions of aggregate price trends, no matter how much you want to shoehorn ABCT into a one-dimensional “supply side” column (the “evil/wrong column”).
John Becker
/ April 18, 2013Price inflation is usually higher at the beginning of busts because, other things equal, fewer goods being produced leads to higher prices. This supply shock that you talk about is exactly what ABCT predicts and what we observe in the real world. Inflation did spike in the first half of 2008 for instance.
The deflation that people associate with deeper recessions is a monetary effect that happens as a result of credit contraction. This typically happens after inflation rises and is usually what people associate with the “panic” phase.
John Becker
/ April 18, 2013Lars,
Here is the illustration of prices going up at the start of the bust before falling due to the inevitable credit contraction that takes place when banks realize they have made bad investments and must rebuild their balance sheets.
http://research.stlouisfed.org/fredgraph.png?g=hD3
Check out Bob Murphy’s article on this subject if you have time. Sorry for the double post.