« How I spent my gap year | Main | Why is Ontario's Government Being So Mean to Its Teachers? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Nick I provided a lengthy reply here [link here NR]. Thanks for answering my questions. JakeS, note that I said that lower rates of interest might lead to higher demand for consumption goods, but not necessarily. And you are quite right that in the current conditions with debt deleveraging processes going on the effects of more expansionary monetary policy are almost nil.

"1. Suppose the price of fertiliser goes down, holding the prices of all the different types of food constant. Will the farmer use more fertiliser? Maybe. Probably yes. He certainly won't use less fertiliser."

It will take me a couple of reads to process this post as a whole, but this stood out. If the price of fertilizer goes down it creates an income effect for the farmer's expenditure as a whole. This could change the composition of what he buys - no need to invoke giffen goods.

Nick: "I disagree. Suppose (for example) we define "the" rate of interest as the one set by the central bank. Now suppose the central bank targets a very high inflation rate. Say 100% inflation. Then the nominal rate of interest set by the central bank will be (say) 105%, while (almost) all real interest rates will be much lower than this."

Okay. Grand. Now, say that the CB holds the rate of interest at 105% and causes the 100% inflation rate -- leaving aside for a moment if they can do this. The the real rate of interest is 5% and, while all the other rates of interest will not be equal to this (I never said that), they would tend toward it.

What you're postulating is an EXOGENOUS shock imposed upon the economy by the CB. The immediate result will be one, in neoclassical terms, of disequilibrium. But in the long-run it is assumed that the economy will return to equilibrium (in your example, at a new price level, so far as I can see).

"But even in that special case of a stationary equilibrium, you still can't say what determines that equilibrium (including the interest rates)..."

I don't recall that I ever did say what "determined" the equilibrium. That is irrelevant to the capital debates, so far as I understand them. What everyone was talking about was an economy in a static state. Because neoclassical economics views the economy as moving toward equilibrium, the Cambridge crowd focused on this equilibrium state in order to show that the equilibrium state itself was incoherent.

I'm not arguing that permanently low interest rates (ZIRP in the (non-Gesellian) limit) eliminates the business cycle. I'm arguing that for a given regime of macrostabilization, lower interest rates will give a higher baseline of economic activity (and greater growth during periods of growth). Because low interest rates (a) reduce the required rate of return for doing stuff, meaning that more stuff will be done (entire sectors of economic activity can become uneconomical at higher or more volatile interest rates) and (b) take less income from net debtors, who are typically the people who spend money (otherwise they wouldn't be net debtors) and give less income to net creditors, who are not inclined to spend money (or they wouldn't be net creditors).

Now, you can argue that (a) will not contribute to aggregate demand, because this investment will simply (for a given state of aggregate demand) crowd out consumption. However, it does contribute to long-run growth, for precisely that reason - at least in economies with structurally insufficient domestic investment (such as the Western(TM) economies of the last three or four decades).

And the point of (b) is that not all income is created equal. Interest income to creditors, in particular, goes into a black hole. Unless you either believe in loanable funds or believe that the creditors have any appreciable inclination to spend their interest income, rather than stash it in bonds or engage in speculative games with it. Whereas interest costs to debtors directly impacts the income of people who are inclined to spend that income.

@Nick Rowe,
causality runs both ways between the quality premium and the total *value* of sheep and the composition of the total volume of sheep.

Does this create practical problems for agricultural economists doing empirical work? Yes, it probably does. [...] But would anybody argue from this that neoclassical economics is incoherent, or that the price premium for meat that is perceived to be of higher quality is somehow illegitimate, and not reflective of relative scarcity? I don't think so.

No, because agricultural economists (usually) don't do silly things like writing down a production function that takes the total capitalization of sheep farmers and attempts to impute a long-run equilibrium value for their revenue or their share of the value added from sheep farming.

But a lot of central bankers do precisely that for the capital plant and profit share of value added.

The problem with capital aggregation in far too many neoclassical models isn't the aggregation per se, it is that (a) the implied causal relationship in the neoclassical model is the wrong way around or (b) there's no correspondence between the aggregates reported by statistical services and the aggregates going by the same name in the models. That's an inclusive 'or' by the way.

Also, as a more general pet peeve, I really think it obscures more than it enlightens to use the term interest rate to describe something that is not a credit instrument. Return to capital, equity or investment, as appropriate, avoids the confusion between access to capital which, in a capitalist economy, is rationed by price and availability of capital, and access to credit which, in a monetary economy, is rationed by political fiat.

@Greg Ransom
If you want to do the microeconomics right, you have do to pure logic of choice of the Robinson Crusoe type WITHOUT categories and goods that only exist in a real, money and credit using economy.

Only if you want to do microeconomics for a species of sentient creatures who in some respects resemble, but are not actually, homo sapiens sapiens.

If you want to do microeconomics properly, you need to start with a good grounding in organization theory, psychology and political science.

- Jake

@ Jake

"Because low interest rates (a) reduce the required rate of return for doing stuff, meaning that more stuff will be done (entire sectors of economic activity can become uneconomical at higher or more volatile interest rates)..."

Totally disagree. "More activity" or "stuff" requires an amount of aggregate demand to facilitate. This is obvious today. If there are not enough buyers -- i.e. too many unemployed or underemployed -- you will not sell your produce and depression will result. No matter of interest rates, they are secondary. Their effectivity falls to nil at a certain point. We're there!

"...and (b) take less income from net debtors, who are typically the people who spend money (otherwise they wouldn't be net debtors) and give less income to net creditors, who are not inclined to spend money (or they wouldn't be net creditors)."

This is true... in theory. But not in practice. See: the Credit Suisse evaluation:


"The side-effect of the Fed’s near-zero interest medicine – the collapse in personal interest income over the last few years. The decline in interest income actually dwarfs estimates of debt service savings. Exhibit 2 compares the evolution of household debt service costs and personal interest income. Both aggregates peaked around $1.4 trn at roughly the same time – the middle of 2008. According to our analysis of Federal Reserve figures, total debt service – which includes mortgage and consumer servicing costs – is down $206bn from the peak. The contraction in interest income amounts to roughly $407bn from its peak, more than double the windfall from lower debt service."

My economics -- and my investment decisions -- involve taking account of actual market conditions. Everything else is nonsense.

A couple of years of following the public debate on economics gives one a healthy skepticism for any analysis coming out of a bank - in particular a Swiss bank.

But even taking their numbers at face value, I'm not prepared to bet money that the marginal propensity to consume or invest (as opposed to gamble on securities) is less than twice as high for net debtors as it is for net creditors. And you shouldn't take their numbers at face value, because bid-ask spreads and bank markups over the policy rate would have increased whether the policy rate was lowered or remained constant. So the spread between interest savings and interest income would still have been there - it might have been smaller, but it's hard to guess how much.

Of course in the current environment, where everybody is deleveraging, a reduction in debt servicing costs doesn't boost demand so much - it does, however, contribute to reducing the duration of the crisis, assuming appropriate fiscal macrostabilization is performed.

- Jake

FWIW: Financial Times commenting

on Bank of England report on who benefits from QE


HT Mark Thoma Economist's View

Frankly, Jake, I think that its VERY believable. Much money made in the US today is financial money. This is because the "spreads" between profits and wages have gotten very "wide". More and more money is being made in the financial sphere and less and less is being made in the productive sphere:


Yes, we can question the "evil" banks so-called "agenda"... but I think that's bullshit. The above is a left-winger and it seems obvious what is going on. Those that deny it are pursuing an ideological agenda and frankly, I have no time for them.

Believe what you want, but don't believe for a moment that it overlaps with reality....

"Illusionist" have you read Bohm-Bawerk?

Hicks learned about his stuff from Hayek, and he admits as a British trained exonomist he never really "got" what Hayek and Bohm-Bawerk doing, couldn't fully escape his training -- Hicks tells us that Hayek was making him think of inputs. Kling before outputs in a process talk place across time. Hicks sought to capture this using British/Marshall and Walras/Pareto methods.

Back at you, "Illusionist", I don't think you understand what the deep issues are here or what a consistent marginalist looks like, or what explanatory role marginalist logic can play, or what the real & deep differences between the Ricardo/classical and the consistent marginalist position really looks like in the domains of valuatational relatikns across time in production goods, and the relation of this pure logic to the explanation of money interest rates in the very different context of the real world.

The Illusionist: "The decline in interest income actually dwarfs estimates of debt service savings. "

That's because

1) Consumers borrow long (30 yrs) and lend short (ignoring retirement accounts)
2) we have a consumer debt crisis and under water mortgage borrowers are unable to renegotiate or move.

Since banks no longer run ALM mismatches, the consumer duration imbalance is actually balanced by retirement accounts, which are in significant surplus (the bond part). But because consumers can't sell those assets, and can't borrow against them, they find themselves massively liquidity constrained. It's a nice trick whereby the financial industry has arranged thing such that consumers are unable to obtain funding by repoing their liquid securities (in registered accounts) and instead are forced to use expensive long term mortgage finance (and policy rate insensitive credit cards).

Since the banks are profiting enormously from the current situation, the only (quick) way to realize the benefit of low rates would be via massive legislated mortgage restructuring. Which doesn't seem likely either at this point.


I agree about the banks not having any ALM mismatches, but I fail to see how there's a consumer duration imbalance. The savings are locked up in retirement accounts, but the 30 year mortgage is not financing the financial assets minus these retirement accounts. It's financing the house, which will presumably last more than 30 years. Where's the imbalance?

In general, if you have a 30 year mortgage at a fixed rate and you have provisioned for mortgage payments out of your income, the decline in house prices is of little concern to you. You care about house prices and mortgages only if you're net short or net long housing, i.e. you have more or less than you need.

@ Ransom

I looked into it. Yes, the Austrians have a typically idiosyncratic (and I would say: woolly) approach to this based on some sort of subjectivism. If this is the theory of interest rates similar to that put forward by Hayek then it is well out of date by now. Sraffa destroyed it in the 1920s and Keynes finished it off in the 1930s.


The house is a real asset. Forget it. I'm talking about fixed rate nominal assets, which are in zero net supply in the economy. For every 30 year fixed rate borrower there's a 30 year fixed rate lender. Since the financial sector is roughly flat, and the corporate and government sector are short, that puts the consumer long fixed rate debt. But there are two factors that prevent most consumers from realizing their MTM gains on their net bond positions:

1) Most of them are actually just short (i.e. they are young or poor and they are debtors). These people actually took mark to market losses on their fixed rate debt exposures.

2) Of the rest, the vast majority hold their bond positions in registered retirement accounts, company pensions or social security. So they have no way of accessing that wealth. You can't take it out, and you can't borrow against it. Though a drop in the term structure of interest rates results in net wealth gains for the consumer sector, they have no way to use it to relieve their liquidity constraints if they are unable to roll their mortgage at reasonable terms.

The problem is partly a balance sheet crisis but also consumer cash flow crisis aggravated by regulatory features of our system of savings which is designed to prevent consumers from efficiently borrowing against their most liquid assets, and which makes the balance sheet problems *look* even more serious than they actually are.

Why did borrowers have to renew mortgages at outrageous spreads in '08/'09 when the government bonds in "their" social security account could be repoed below 0%? It's a very elaborate financial Rube Goldberg. But it's still a scam.

In Canada, you can't access your employer-sponsored retirement account or your CPP-QPP account but your RRSP ( registered retirement savings plan)can be accessed for home-buying and higher-education retraining. Was the decision based on the realization that the problem you raised is very real? I don't know but sure thre was a great media and public opinion pressure. Though the practice is sometimes discouraged on the fear that it won't be paid back and that you are stopping the accumulation of assets.
Unfortunately, for most people, the only moment when they access their RRSP is at bankrupcy proceedings , when it's too late to do any good.

Jacques: "Though the practice is sometimes discouraged on the fear that it won't be paid back and that you are stopping the accumulation of assets"

I think that's largely "concern trolling" by the financial industry. If we had wanted to address that issue, we could have given people an easy way to access a mortgage from their RRSP at the same terms as (and pari passu to) their bank mortgage. That way it can only be a substitute for, rather than an addition to, other debt. This also precludes any argument for the $25K limit. I would have loved to give a 30-year mortgage to myself (with outrageous prepayment penalties!) in my RRSP 15 years ago. For some reason, that solution wasn't proposed by the (deeply concerned) financial industry.

Not that I'm mainly advocating that, though a bank mortgage provides an excellent return and lending to yourself is a really excellent risk. But beyond that, we need to be able to access *all* of our superannuation assets as collateral. Securities are way better collateral than houses.


The Godley/Lavoie models indicate that lowering interest rates provides a short-term boost to demand, but creates a long-term drag due to lost interest income.

id need to see that carefully (any references?)

for one, there are distributinal issues involved (welath inequality is higher than income inequaltiy). only a very limited fraction of the population would see their income increased this way, but financial costs would increase for everybody. I cant see why the income increase effect would dominate the cost increase effect.

Anyhoo, the way you expressed it, it would contradict the very logic of inflation targeting,. what would stop it from generating an inflatinary process like so

higher interests -> higher income -> higher demand -> higher prices -> higher interests

"TheIllusionist" -- beg the question much?

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad