If I see one more model of secular stagnation and negative equilibrium real interest rates, that does not include land….I'm going to throw a real wobbly.
What is it with you townies? Have you never looked out of the window, when you fly (do you ever drive?) from one city to another, and wondered about all that stuff you see out there? It's called "land". It grows food, that you eat. And that land is valuable stuff, and there's a lot of it, and it can last a very long time, and it pays rent (or owner-equivalent rent). And if the rent on that land is strictly positive (which it is), and if the price of that land is finite (which it is), then the rate of interest you get by dividing that annual rent on land by the price of land is going to be strictly positive. And that's a real rate of interest, because land is real stuff, and what it produces is real stuff too.
So when you go to a helluva lot of trouble to build a model with a negative equilibrium real rate of interest, and it's a very fancy complicated model, but it totally ignores land, I really wonder where you are coming from. Actually I don't wonder. I know where you are coming from. You are coming from the town, or the big city, where you can easily forget about land. But even then: you know that stuff your house or condo is built on? That's called "land" too.
It didn't used to be this way. The Physiocrats, Malthus, Ricardo, didn't ignore land. Land was central to their models. Now I know that whether or not you should include something in your model depends on what you want your model for. And that it is perfectly OK to ignore land in some models. But if you have a model of secular stagnation and negative real interest rates that ignores land, well, your model is just hopelessly and fatally wrong. Dead wrong. It's just not grounded.
You don't need fiscal policy to "solve" the problem of permanently negative equilibrium real interest rates. You need land. You cityfolk need to pay attention to land. Maybe you haven't noticed that land prices have been rising recently, worldwide? Think that might have something to do with low equilibrium real interest rates and secular stagnation? Think that land might maybe be a savings vehicle in an overlapping generations model? Think that the bursting of the housing land "bubble" might maybe have been just a conjunctural hiccup in a much bigger secular story?
I read the abstract and introduction of the Eggertsson and Mehrotra paper (pdf) (HT Paul Krugman), got the gist of it, but saw no mention of "land". Then I skipped to the references, saw no reference to Stefan Homburg's classic paper, so I stopped reading. All that effort to build that fancy model, all for nothing. (Here is Stefan Homburg's latest working paper (pdf) on the subject, which shows that land is macroeconomically as important as capital, and more important than public debt.)
Lefties sometimes complain about a class bias in research. I'm going to complain about an urban bias in research. Some of you academics should be rusticated for a bit, until you learn better.
Worthwhile Canadian Initiative 
“If I see one more model of secular stagnation and negative equilibrium real interest rates, that does not include land….I’m going to throw a real wobbly.”
How man of these landless stagnation models have you seen? Thanks for “rusticate” … and “wobbly”… new to me.
I had the same thought yesterday–about whether land could provide an escape from secular stagnation–and pulled Martin Bailey’s old (1972?) macro textbook down from my shelf. In response to some papers about secular stagnation from the 1930s, Bailey argued that (long-term) equilibrium real interest rates couldn’t be negative because there were various investment opportunities that would overwhelm the stock of savings at very low real rates. As one example, he considered the payoff of expanding the US land mass into the Gulf of Mexico. I wonder if an updated version of Bailey’s calculations would still support his argument (assuming of course that such a large landfill could pass environmental approvals….).
Tom: let’s ask the question the other way: how many secular stagnation models have I seen which do include land? Only Stefan Homburg’s.
Angelo: lovely example! So Bailey still remembered land, in 1972 or so. And that example reminds me of a very similar example my high school economics teacher, Mr (Geoff?) Best told us kids in about the same year: if the real rate of interest fell to 0%, it would be profitable to bulldoze the Rocky Mountains flat and convert them into farmland. But even if we can’t create more land, the existing stock of land, used as a savings vehicle in an OLG model, would rule out a permanently negative real interest rate. (OK, you could maybe get it to go negative again with enough modelling twists, but the modellers should concentrate their attention on those twists, before they talk about borrowing constraints etc.).
One consideration regarding land ( and yes, what are economists thinking to ignore it? Do they live in Laputa?) is that ownership of land is not necessary for control of land.
If that ownership is not defended by, as always in the end, some branch of the state, then that ownership is precarious. Conversely, if the wealthy need a chunk of land, they need not buy it up front or outright — they merely need to exert control over the defensive capabilities of the owner or the state.
In fact, I am convinced that one defining characteristic of the rich is their invariable tendency to spot something they like or need and say, “Hey, that’s nice, I will take it.” They don’t need to own everything, they just need enough influence to acquire the use of it when they wish, and shed the responsibility for it when it becomes a burden.
So, the gold mine is theirs so long as it yields gold, but not when the tailing ponds (read “poison lakes”) burst their bonds and wreck a watershed.
Noni
“you know that stuff your house or condo is built on? That’s called “land” too.”
FYI most economists tend to live in little castles that float in the air, or on houseboats.
Nick,
The city folk prefer talking about robots.
Not sure what’s so special about land, as compared to any other storable commodity. Sure it pays positive rents and has an essentially infinite shelf life, so in some sense the “real” return is strictly positive, and it may be unique in this respect, but I don’t think there’s anything special about the line that divides negative from positive returns. (It’s not like anyone is targeting exactly zero inflation.) Like any other storable commodity, land is limited in its uses: you can’t costlessly (or at a cost that’s bounded low) convert land into clothing or health care or entertainment. So land is risky: there’s no guarantee on the real price of land (or its direct services) in terms of the generic commodity. So land carries a risk premium. We don’t know exactly how high that risk premium is. Maybe it’s actually lower than the risk premium associated with money, in which case secular stagnation is impossible. But maybe it’s much higher. Not implausible that it’s high enough to render relevant models of secular stagnation that don’t include land.
Interesting comment, Nick. Did you see my post a few weeks back (http://monetaryreflections.blogspot.co.uk/2014/03/samuelsons-consumption-loan-model-with.html)? I described a structure very similar to theirs, but based it on debt incurred to acquire land, rather than to fund consumption. I didn’t talk about interest rates or secular stagnation, as I was looking at other issues, but I was implicitly assuming some dis-utility to holding land in order to justify having a zero interest rate (as described in my immediately succeeding post).
This is an excellent article. But I think that this selective blindness to land does more – as shown in this brilliant example by Karl Smith as a reaction on latest piece from Piketty: http://ftalphaville.ft.com/2014/02/03/1759582/piketty-and-the-case-for-land-capital/
I think I agree with Andy Harless, a secular stagnation needs to include price uncertainty.
I might have this wrong, but isn’t the secular stagnation idea and permanently negative real interest rates contingent on a perpetual negative output growth, with a reflecting real price depreciation making the realized investment real yield negative.
e.g. A country has a current population of 127 million people, who all demand real resources such as land etc. keeping current prices high. Assume that that same country for demographic reasons is expected to half its population and output in 50 years. The price of land will presumably depreciate over that period, buying less goods and services at an eventual sale, making the real yield from investing in land negative.
“And if the rent on that land is strictly positive (which it is), and if the price of that land is finite (which it is), then the rate of interest you get by dividing that annual rent on land by the price of land is going to be strictly positive.”
Maybe. I think this leaves out a dimension, however. Let’s say short-term real rates on financial assets are negative. Land, being finite in price, gives you a positive current income on your investment. So you buy land, collect income at say 2% per annum on your initial investment. Great. But that 2% is not your rate of return, unless you are able to sell the land for the same price for which you purchased it. And if rates are higher, it is likely the price of land is now lower.
If short term rates are negative, and expected to stay negative forever, then capital will flow into long-lived assets like land to get a superior real return, and short term rates will not stay negative. But things get more complicated if you start to think of a term structure to rates.
Brillant piece. And the townies should also remember this: The value of an appartment consists of the value of the structure plus the value of land.
On the average, the land value is about one half of the total appartment value (a quarter at the moment the appartment is built, rising to one hundred percent as the structure deteriorates).
Hence, townies: Half of the rent you pay for your apartment is land rent! Do you think this is negligible?
Perhaps these effects can also be explained with a liquidity argument?
Land is a near-perfect store of real value, by your arguments above: if you want to preserve the value of money infinitely, buy land. At the same time, it’s a rotten liquidity instrument, in that changing the use of land has large transition costs (a farmer can’t exactly cash out land to developers, only to buy it back and bulldoze the houses after the price crashes).
So as the liquidity utility of a dollar tends towards zero, it becomes more and more worthwhile to seek a savings vehicle that better preserves value.
Maybe traditional, first-year economics is wholly right about monetary policy: we are seeing a classic hot-potato inflationary effect, but only in savings vehicles because people collectively don’t value the marginal liquidity of a new currency unit. (Also the flip side of the permanent income hypothesis, in that agents don’t feel that their permanent income is liquidity-constrained and so don’t have unfulfilled desires for consumption.)
Nick, just so you know, this post doesn’t show up under “Posts by author” which is how I usually keep track of your latest.
Majormax,
I agree. In fact I would argue that liquidity problems associated with land can be broken into a couple subgroups:
1. Land suffers from divisibility issues – farm land (not so much), residential housing (absolutely)
2. Land suffers from portability issues – land can’t be easily relocated
Also, land suffers from subjectivity issues – there are no two identical pieces of land
“You don’t need fiscal policy to “solve” the problem of permanently negative equilibrium real interest rates. You need land.”
So, you’re advocating that the U.S. conquer Canada to get more land to avoid secular stagnation?
Land is even more important and valuable in urban areas. I don’t view secular stagnation as a permanent state of affairs though, but a transitory uncertain one. Land becomes more valuable during such times but its value and return less certain. The longer it persists, flattening or even falling population can drive it in reverse where it may be the best available option but not necessarily a good one. People aren’t walking away from land, yet or still, though they were doing so, via foreclosures, just a few years ago. Up again, but still below where it was, above or below where it will be at some future date?
Lord, “Land is even more important and valuable in urban areas.” I was thinking the same thing.
Andy: OK, 0% real interest isn’t a magic number. What’s really at stake is dynamic inefficiency, which is only the same as r=0% in a stationary model with no growth.
There’s nothing special about land, except that any infinitely durable good in fixed supply that pays rents can beat Samuelsonian “money”. Stefan Homburg does say his result generalises to a multi-good stochastic environment. I’m not sure I fully get the proof, but I think the intuition is this: what would you rather own as a savings vehicle: a bit of paper that might pay capital gains; a bit of paper (the deeds to land) that might pay capital gains, and might also pay some stochastic dividend? If the latter, then land dominates Samuelsonian “money”.
Now money is more liquid than land, and its price is short-run sticky, so money might dominate land in the short run. But for long run saving, those two considerations shouldn’t matter.
Nick E: interesting. But assuming an equilibrium where land has zero marginal utility looks problematic to me. that would mean land rents would be zero. My preference would be for your bank to own the land directly, then rent it out to households to play on. But with positive rents on land, the real rate of interest would be positive.
JV: Thanks! Unfortunately I can’t read the FT piece you link to. Stefan Homburg was following similar lines in an email to me.
Mikael: “I might have this wrong, but isn’t the secular stagnation idea and permanently negative real interest rates contingent on a perpetual negative output growth, with a reflecting real price depreciation making the realized investment real yield negative.”
For a stationary economy, with constant growth and real interest rates, the key question is whether the real interest rate is greater or less than the real growth rate. If r < g, you get dynamic inefficiency in an economy without land, and a Ponzi scheme can be sustainable and make all generations better off. But land dominates a Ponzi scheme.
louis: true. You need a more complicated model when r varies over time, but the result should generalise. My statement was just an intuitive idea, rather than a formal proof. See the linked papers for something more rigorous.
Herbert: thanks! (And you of course deserve credit, for tipping me off a year or so back.)
Majromax: Yep, liquidity matters, for shorter holding periods. That’s why we hold money even when it is rate of return dominated. But it should matter a lot less at longer horizons, like saving for retirement. Plus, pension plans, and banks, could hold land on the asset side, in exchange for more liquid liabilities.
Tom: thanks. Fixed it.
Redwood: nope. The US already has lots of land. And all you need is some for this effect to work; you don’t need more. Plus you could also buy foreign land.
Lord: the more time that passes since the “house” price crash, the more I wonder if it wasn’t just a temporary aberration, and only in some countries.
Andy: let me make a minimalist point: at a minimum, one would want people who model secular stagnation and say there is a role for fiscal policy to include land in the model and show how their claim is still valid despite the existence of land. Simply including land in the Eggertsson Mehrotra model would overturn their results. It’s not as though land is some tiny thing we can ignore. Land is BIG. (At the very end of their paper, they talk about including capital in a future paper, but no mention of land.)
“Lefties sometimes complain about a class bias in research. I’m going to complain about an urban bias in research. Some of you academics should be rusticated for a bit, until you learn better.”
I’m excited that we’re entering the Maoist phase of Nick Rowe’s intellectual development.
(Yes, I know that rustication isn’t AS bad as what Mao did. Also- I like to see Durham in a list with Cambridge and Oxford!)
W Peden: Yep, the thought of mentioning Mao did cross my mind, but I decided against!
Durham’s pretty old, IIRC. It wouldn’t even let me in, when I applied through UCCA.
W. Peden,
“I’m excited that we’re entering the Maoist phase of Nick Rowe’s intellectual development.”
That is hilarious!
Durham,
It’s only from the 19th century (1832) and it’s actually the youngest university I’ve attended. By contrast, I think that Edinburgh University is the youngest of the main Scottish universities, and it was established 1583. Still, most of the town itself is older, as is the cathedral that dominates the centre.
Most importantly, being in the north of England, it’s a cheap and friendly and beautiful little place.
Andy Harless, I think W. Peden has come up with the perfect title for your rebuttal post to Nick here:
“The Maoist phase of Nick Rowe’s intellectual development”
Nick, I think you can get past the paywall for Karl Smith’s post by going through the link on Google’s search results for “Piketty and the case for land capital”.
Land is technically producible on a systematic scale, nowadays. Holland, Macau…
Rather, the thing about land is that a lot of its value appears to be positional, in a literal fashion.
Couldn’t the price for land become infinite? Why should a corporation or a family (any infinitely lived agent) sell land for less if there is no other assets having a positive rate?
And shouldn’t especially those living in cities be attentive to exploding land value? Isn’t a lot of Piketty’s patrimonial capitalism about land?
Nick,
As someone who grew up on a 200 acre farm with crops (for townies: things that grow in the dirt) and animals (for townies: creatures with four legs) and who later became a Mortgage Manager in the 4th largest branch of the third largest Canadian bank lending millions of mortgage dollars on what we – and many others called – hold on – REAL estate (as distinct from non-real estate or all the other stuff), your post brought a smile to my face.
But the existential importance of land was captured far more eloquently by Gerald O’Hara in Gone with the Wind: “Do you mean to tell me, Katie Scarlett O’Hara, that Tara, that land doesn’t mean anything to you? Why, land is the only thing in the world worth workin’ for, worth fightin’ for, worth dyin’ for, because it’s the only thing that lasts”.
I just finished re-watching the magnificent 1980 BBC series Churchill: The Wilderness Years. It seems you are becoming positively Winstonian against the Chamberlains (or MacDonalds) of economics?
Japan? Falling land prices?
I know you only said they hadn’t, and should, not that they couldn’t model land prices. But, if you tried, wouldn’t falling land prices create effective negative interest rates for as long as prices were falling.
Now why would land prices fall? When relative yield is high?
Well, land is long duration so its yield won’t go to zero (limit) for the same reason that long bonds won’t (option value) so there is some positive yield that is too low in all circumstances.
Presumably land prices in Japan had such miniscule yields that prices could fall by half.
While prices fell, the real yield on land was negative.
Being very long in the property market with an agricultural farm and a toronto condo, I understand the virtues of property as well as its costs. However
1) property wealth is badly distributed. Net housing wealth is somewhat more broadly distributed but is not really a fungible financial asset because of its shelter services. Contrary to their marketing, the “income from your house” programs are extremely expensive.
http://www.cagle.com/2014/03/canadian-middle-class/ is an Ingrid Rice cartoon that came out after the initial data release from the SFS
2) Compared to the volume outstanding of financial and other assets, land is a trivial part of the picture and would affect the expectations (that is what really matters) of a very narrow share of the population. While your point about its exclusion is nice, in the modern world, it is irrelevant in the scale of decision making of the broad economy.
Excellent post, Nick.
Land market failure leads to capital market failure. I’m surprised you think secular stagnation economists don’t recognize this and put it into their models because they’re urbanites. Cities are where distortions in the land market are most obvious. I see the land market failure in New York City more than anywhere else, like the one-story buildings next to high-rises.
The economist who keeps land in the model and relates land monopoly to capital market failure is Mason Gaffney. He wrote an essay explaining this titled, “Great Expectations: How credit markets twist the allocation and distribution of land,” which can be found here: http://schalkenbach.org/rsf-2/great-expectations/
Cannot the real return on land be negative? Falling market value, depreciation, maintenance, and other expenses could add up to more than your income or use value derived from the property.
Land is also something that you can’t make fungible, can’t move, and really can’t hide. Land is something that attracts coercion. Land requires a defense; and that, from the taxman.
In short, land can have a negative value. It can put you in jail or cost you your life.
The Maoist phase, indeed. Worse, see what happened to land owners under Stalin. Or look to the mansions of Newport and Long Island.
BenK, I like the comment.
“Land is something that attracts coercion.”
Not that you’re doing this, but I hear from certain “homesteading” advocate commentators all the time, that it attracts “jack booted government thuggery from sociopaths with guns who initiate violence.”:
http://www.themoneyillusion.com/?p=26468#comment-327455
😀
W. Peden: Just for history: when I studied in the UK (long ago) we were told that universities were grouped in “White stone” (Oxbridge)(for the old ruling class), “red brick” (for the 19th century industrial bourgeoisie, Durham being the oldest) and “grey cement” (postwar educationnal boom for the working classes). Is this expression still current? And is there a name for the fourth generation(polytechnic turned universities?)
My whirlwind tour took me to East Anglia (Norwich), Cardiff, York and Galashiels (now Borders)
Isn’t the return to land the sum of rent and the change in the price of the asset? Thus, the real interest rate (return, whatever) can easily be negative when the asset value falls at a rate greater than the rental yield?
If you are not including the total returns to an asset, how do you determine what is ‘interest’ and what is ‘something else’?
Jacques Giguère,
I’ve never heard that expression, except “red brick”. Since it misses out the four ancient Scottish universities (until the foundation of the University of Durham, Scottish universities outnumbered English universities 2 to 1) I have no doubt that it was a genuine expression. Perhaps one could add “grey stone” for the Ancient Universities, given our love of building with stone. I’ve been fortunate enough to have contact with three out of the four types of university: two pre-19th century universities (undergraduate at Edinburgh, masters at Cambridge) one “red brick” (PhD at Durham) and one “grey cement” (teaching at Stirling, Nick Rowe’s alma mater). One of the things that the UK still does well is higher education, in spite of the many bad interventions of governments of all colours.
Also, as far as degree-level education for the working classes goes, the really big revolution was the expansion under New Labour with the introduction of tuition fees. However, I imagine that this massive social change wasn’t much discussed before it happened…
I live in an agricultural area, and the value of land is often negative here. You have to pay tax on land. Land prices rise and fall in response to rumors of development, availability of credit, urban food attitudes, and commodity prices. Land prices may rise eventually, but this is on a scale longer than the lives of most individuals.
Also, a lot of the land is timberland, so its value drops after a harvest, and is unlikely to recover for another 40-80 years. Sure, timber prices may be higher in 2064, but your tax bill is due tomorrow.
One note is that Summer’s definition of stagnation is “real rate consistent with full employment is negative”. That is not inconsistent with positive yields on land, capital, or anything else. Assume we have an economy of consumers, land owners, and contract farm workers (migrant labor): the land owners will only hire farm workers while it is profitable to sell the output to consumers. If the the marginal output from hiring workers is decreasing and hits zero before all labor has been hired then we have stagnation.
Also I would note that inflation rates vary from product to product. If the rate of inflation on crops is lower than on things the land owner wants (education from Canadian economists or Singapore software guys?) then the effective return on land ownership might be negative even if there is positive nominal return.
The problem is while land, at least select land, will generally have positive value, it doesn’t mean it will have a positive return. Since land prices represent the capitalized returns expected, lower cap rates will increase the value, but lower expected returns will decrease it. If near expectations fall more than cap rates, the near return will be negative and prices will fall. It then becomes a competition of how much they will fall, how fast and how much cap rates will follow, how soon expectations may turn, how fast and how much they may improve, how much risk you are willing to take, and how much reward you want for it. A world in which all returns are ground to less than that of land is one in which the return of land is also ground to near dust and the only reason to buy land is you expect those returns to turn, otherwise money remains the superior ‘investment’.
Nick: no comment on Noni’s comment? I think it’s very insightful.
We’ve seen an awful lot of what I can only call land theft lately, with the mortgage fraud / robosigning crisis. If you look back in history, of course, it consists mostly of land theft, including nearly the entire occupation of Canada and the US (stolen from the Native Americans, and they’re still fighting to get paid for it in many cases).
Returns on land may end up being nothing more than returns on ability to throw power around in cases like this. What do you do without the rule of law?
Nick, thank you for your illuminating blog and for this intelligent post. However, I don’t see how the existence of land changes the argument about secular stagnation. My understanding of Larry Summers’ sec stag hypothesis is that the theoretical short-term risk-free rate consistent with full employment is theoretically negative–and could remain so for some time to come. In other words, even with a 0.25% nominal federal funds rate (in the U.S.), there would still be more desired savings at full employment than there is productive investment. The Eggertsson and Mehrotra paper is just simplified model with one interest rate to show how this can occur without a self correcting mechanism when there is a slowdown in population growth. I think the existence of land theorem is saying that because land theoretically always has a positive return (unless land values are falling), there could never be a shortage of profitable investment opportunities regardless of how much saving there is. Did I understand that correctly? Maybe I misunderstood what you all are saying about land. But if I did get it right, this doesn’t seem, to me, to pass the sniff test. Even if everyone only consumed 10% of their income, you could still maintain full employment by productively investing all the savings in land?! Why haven’t people woken up to this panacea (land) after nearly 6 years of sitting on money? My apologies if I am totally missing the boat regarding what you are saying about land; please correct me if this is the case.
Nick, I’m late and haven’t read all the comments so sorry if this is a repeat but:
“Then even though land has a positive marginal product, the expected
return from investing in land can, in principle, be negative.
This is a highly stylized example, which begs many questions.
Nevertheless, it at least establishes that a liquidity trap can occur despite
the existence of productive investment projects.”
That’s from:
Click to access 1998b_bpea_krugman_dominquez_rogoff.pdf
Krugman explicitly considers land and explicitly shows it needn’t change anything about the argument, I don’t think your counter example works.
I think this is all missing the point.
I think the argument here is really about what the “equilibrium” rate of interest actually means. And I think Andy Harless has this right – this is all about perceived risk. The issue is not whether something gives a particular positive rate of return, it is whether that positive rate of return compensates for the perceived risks involved or not. Farms may on average have a positive rate of return, but plenty of farmers still go broke.
reason, and Andy,
yes risk is a key feature and also addressed by Krugman, another quote from the BPEA paper:
If the equity premium is as high as the historic U.S. average, the economy could find itself in a liquidity trap even if the rate of return on physical capital is as high as 5 or 6 percent.
Nick: Sorry for later response. As per the latest Piketty book “Capital in the Twenty-First Century” he has this nice graph somewhere in there: http://imgur.com/Hkhr4rM
Which basically confirms your point. The land is back! Only some reviewers used the same graph against some of Pikkety’s conclusions especially those pertaining taxes. The optimal tax of land is different compared to optimal tax on other capital goods like machines. It seems that your advice is valuable for all economists, not only those thinking about secular stagnation.
Guys: it’s exam time, (plus I’m burning out a bit), so I’m not responding to comments as much as I should.
But I think we need to distinguish between: a short term liquidity trap/ZLB problem (for which e.g. a higher inflation target or NGDPLT would be the cure); a secular long term problem of dynamic inefficiency, for which fiscal policy is claimed to be the cure. This post, and the role of land, is really about that second problem.
I would encourage you to read Stefan Homburg’s working paper linked above to get the formal analysis, but here’s the gist of it: if there is dynamic inefficiency (and over-investment in capital as in Diamond 65), then a government Ponzi debt is the solution, where the government increases the level of debt, and borrows forever to pay the interest on the debt. (With dynamic inefficiency, it is efficient for government borrowing to crowd out investment in capital.) But land is at least as good as Ponzi debt, because it pays rent. Land already solves the problem that Ponzi debt is supposed to solve. Samuelson 1958 already recognised this, when he said he needed to assume that land is “superabundant”.