Is aging deflationary?

Hideki Konishi, Kozo Ueda, and Mitsuru Katagiri presented a few months ago a paper titled "Aging and deflation from a fiscal perspective." Here are the slides.

The authors build a model that combines overlapping generations, the fiscal theory of price determination, and political considerations to analyze how the price level changes with fertility and longevity.

The simplified version of the model, in section two of the paper, assumes that taxes are exogenous. This simple version, nevertheless, is enough to produce a key result:

"Aging is deflationary when caused by an increase in longevity, but inflationary when caused by a decline in birth rate." 

The reason is political considerations:

"If the birth rate declines, the resultant contraction in the tax base reduces the fiscal surplus. The government is then inclined to maintain its solvency partly by generating inflation at the cost of the older generation's well-being and partly by making the younger generation pay more taxes. In contrast, if the life expectancy increases and older persons survive longer than expected, they might face a shortage of savings for their retirement period. The government then, led by the strengthened political influence of the older generation, attempts to suppress inflation and increase the real value of the government bonds held by the older generation."

Japan has experienced both unexpected declines in fertility and unexpected increases in longevity. The deflationary effect of higher longevity, however, dominated. The authors' simulations of the model show that Japan's aging depressed inflation by 0.6 percentage points annually.

Another key result, which comes from the fiscal theory of price determination, is that our children don't pay for our deficits. The government debt at the beginning of each period is fixed in nominal terms. Today's price level changes to equate the real value of today's debt with the present value of future deficits.

A corollary of this result is that tomorrow's fiscal policy is not constrained by today's level of debt or fiscal policy. Governments are unencumbered by the deficits of their predecessors in office.

A second corollary of the fiscal independence result is that governments don't have an incentive to strategically accumulate debt. In some political economy models of fiscal policy, a government can tie the hands of a successor it dislikes, by raising debt. If the price level, however, adjusts every year to fulfill the government's inter-temporal budget constraint, strategic debt accumulation doesn't happen.

I thought this paper was a refreshing way of looking at the link between deflation and aging.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

[The following list was edited on March 17, 2015.]

Other recent papers (post 2000) about this topic, in chronological order:

Fair, R. C. and K. Dominguez (1991), “Effects of the changing U.S. age distribution on macroeconomic equations”, American Economic Review, 81(5), pp 1276–94.

The effects of the changing U.S. age distribution on various macroeconomic equations are examined in this paper. The equations include consumption, money demand, housing investment, and labor force participation equations. Seven age groups are analyzed: 16-19, 20-24, 25-29, 30-39, 40- 54, 55-64, and 65+. There seems to be enough variance in the age distribution data to allow reasonably precise estimates of the effects of a number of age categories on the macro variables. The results show that, other things being equal, age groups 30-39 and 40-54 consume less than average, invest less in housing than average, and demand more money than average. Age group 55-64 consumes more and demands more money. If these estimates are right, they imply, other things being equal, that consumption and housing investment will be negatively affected in the future as more and more baby boomers enter the 30-54 age group. The demand for money will be positively affected. If, as Easterlin argues, the average wage that an age group faces is negatively affected by the percent of the population in that group, then the labor force participation rate of a group should depend on the relative size of the group. If the substitution effect dominates, people in a large group should work less than average, and if the income effect dominates, they should work more than average. The results indicate that the substitution effect dominates for women 25-54 and that the income effect dominates for men 25-54.


Lindh, T. and B. Malmberg (1998), “Age structure and inflation: A Wicksellian interpretation of the OECD data”, Journal of Economic Behavior and Organization, 36(1), pp 19-37.

Wicksell's cumulative inflation process is founded on the separation of investment and saving decisions. The demographic age structure influences the aggregate of both these decisions, and therefore should be one of the determinants behind the inflation processes. We study annual OECD data 1960–1994 using age variables to explain inflation. Panel estimations of a reduced form inflation-age model show a robust correlation consistent with the hypothesis that increases in the population of net savers dampen inflation while especially the younger retirees fan inflation as they start consuming out of accumulated pension claims. This pattern is expected from life-cycle saving but could also be due to age effects on budget deficits or on money demand. Our results are potentially important for inflation forecasts and monetary policy.


Lindh, T. and B. Malmberg (2000), “Can age structure forecast inflation trends?”, Journal of Economics and Business, 52, pp 31–49.

The demographic age structure influences the aggregate of individual economic decisions. Standard macroeconomic models imply that inflation pressure will covary with the age distribution unless accommodated by monetary policy. We estimate the relation between inflation and age structure on annual OECD data 1960–1994 for 20 countries. The result is an age pattern of inflation effects consistent with the hypothesis that increases in the population of net savers dampen inflation, whereas especially the younger retirees fan inflation as they start consuming out of accumulated pension claims. This can be explained, for example, with life-cycle saving behavior combined with a cumulative process of inflation, but other mechanisms are also consistent with the results. In any case, the results suggest that demographic projections may be useful for long- and medium-term inflation forecasts. Forecasts from our panel model catch the general downward trend in OECD inflation in the 1990s. However, useful forecasts for individual countries need to incorporate more country-specific information.

Bullard J., C. Garriga and C. J. Walker (2012), “Demographics, Redistribution, and Optimal Inflation,” Federal Reserve Bank of St. Louis Review, November/December 2012, 94(6), pp. 419–39.

The authors study the interaction among population demographics, the desire for intergenerational redistribution of resources in the economy, and the optimal inflation rate in a deterministic life cycle economy with capital. Young cohorts initially have no assets and wages are the main source of income; these cohorts prefer relatively low real interest rates, relatively high wages, and relatively high rates of inflation. Older cohorts work less and prefer higher rates of return from their savings, relatively low wages, and relatively low inflation. In the absence of intergenerational redistribution through lump-sum taxes and transfers, the constrained efficient competitive equilibrium requires optimal distortions on relative prices. The authors’ model allows the social planner to use inflation/deflation to try to achieve the optimal distortions. In the model economy, changes in the population structure are interpreted as the ability of a particular cohort to influence the redistributive policy. When older cohorts have more influence on the redistributive policy, the economy has a relatively low steady-state level of capital and a relatively low steady-state rate of inflation. The opposite happens when young cohorts have more control of policy. These results suggest that aging population structures, such as those in Japan, may contribute to observed low rates of inflation or even deflation.

Anderson, D., D. Botman and B. Hunt (2014), ”Is Japan’s Population Aging Deflationary?” IMF Working Paper 14/139, August.

Japan has the most rapidly aging population in the world. This affects growth and fiscal sustainability, but the potential impact on inflation has been studied less. We use the IMF’s Global Integrated Fiscal and Monetary Model (GIMF) and find substantial deflationary pressures from aging, mainly from declining growth and falling land prices. Dissaving by the elderly makes matters worse as it leads to real exchange rate appreciation from the repatriation of foreign assets. The deflationary effects from aging are magnified by the large fiscal consolidation need. Many of these factors will beset other advanced countries as well, but we find that deflation risk from aging is not inevitable as ambitious structural reforms and an aggressive monetary policy reaction can provide the offset.


Yoon, J.-W., J. Kim and J. Lee (2014), “Impact of Demographic Changes on Inflation and the Macroeconomy” IMF Working Paper 14/210 November.

The ongoing demographic changes will bring about a substantial shift in the size and the age composition of the population, which will have significant impact on the global economy. Despite potentially grave consequences, demographic changes usually do not take center stage in many macroeconomic policy discussions or debates. This paper illustrates how demographic variables move over time and analyzes how they influence macroeconomic variables such as economic growth, inflation, savings and investment, and fiscal balances, from an empirical perspective. Based on empirical findings—particularly regarding inflation—we discuss their implications on macroeconomic policies, including monetary policy. We also highlight the need to consider the interactions between population dynamics and macroeconomic variables in macroeconomic policy decisions.

Juselius, M. and Takáts, E. (2015), “Can Demography Affect Inflation and Monetary Policy?” BIS Working Paper 485, February.

Several countries are concurrently experiencing historically low inflation rates and ageing populations. Is there a connection, as recently suggested by some senior central bankers? We undertake a comprehensive test of this hypothesis in a panel of 22 countries over the 1955-2010 period. We find a stable and significant correlation between demography and low-frequency inflation. In particular, a larger share of dependents (ie young and old) is correlated with higher inflation, while a larger share of working age cohorts is correlated with lower inflation. The results are robust to different country samples, time periods, control variables and estimation techniques. We also find a significant, albeit unstable, relationship between demography and monetary policy.

Fear depression, not deflation

Following up on a blog post by David Andolfatto, I checked on the deflationary experiences of seven countries during the 19th and 20th centuries.

The mainstream commentary these days is that deflation causes (or at least is associated with) declining real economic activity. Here's an example of this type of narrative:

Plenty of people are alarmed by the prospect of deflation, which can snuff out growth by making consumers reluctant to spend and companies unwilling to invest.

David's main point is that that's not always the case. Both the U.S. after the Civil War, and Japan since 2009, experienced declining price levels, but real GDP per capita rose.

I have put together a dataset of price indexes and growth between 1828 and 2006 for Australia, France, Netherlands, Spain, Sweden, U.S., and U.K. (Many thanks to the sources, especially measuringworth.com, and the International Institute of Social History, for distributing the data for free!)

It's relatively well known that prices were roughly steady in the 19th century, although income per capita rose quite a lot. The inflection point seems to be the Great War, but that's something I learned only after looking at the data.

See these charts and table:

This seems hard to reconcile with the conventional association between "lack of inflation" and "lack of growth."

Taking five-year growth rates, the association between inflation and real growth is hard to spot (the picture is very similar with one-year growth rates):

Finding "big" deflations

Ok, maybe you're thinking: 19th-century deflation was slow and steady. Under that type of deflation, declining prices are presumably expected, which should be less damaging than rapid, unexpected deflation.

To which I reply: But slow-and-steady is the kind of deflation that we're contemplating these days, right?

How about abrupt, big deflations? I don't know of a definition of "big" deflation, but I think most people have the American Great Depression in mind.

The U.S. price index bottomed in 1933, at which time the five-year, annualized inflation rate was -5.4%. So I have looked for deflations of this size across countries. When I find one, for a given country, I rule that a "big" deflation episode occurred, starting with the first year in which one-year inflation was negative, and ending with the first year in which prices rose (even if the five-year inflation rate was no longer below -5%). If the string of price declines is interrupted for only one year, I consider the episode of deflation was unbroken; if the string stops for two or more years (and five-year inflation eventually falls below -5%), then a new episode begins.

Using this admittedly ad hoc process, I find the following episodes of "big" deflation:

The table shows that, for example, Australian prices fell at a compounded rate of 3.6% a year between 1836 and 1851, for a total decline of 43%. Real GDP per capita went up by 5% a year.

Except for the deflations in the late 1920/early 1930s, output per capita didn't fall during deflations! It has been possible, and in fact frequent, to have rising real economic activity and falling prices, even with rapid deflation.

At the very least, I think we should agree that not all deflations are created equal. Only a small minority is associated with persistently declining real activity. And those experiences, over the last two centuries, have all happened during a peculiar period (the Great Depression).

What really hurts, then, is not deflation, but a depression, which is not what most people expect for the years ahead.

Data sources:

Real GDP per capita growth:

Maddison Project Database.

Inflation:

Australia: Diane Hutchinson
France: CGEDD
Netherlands: International Institute of Social History
Spain: Rafael Barquín Gil, Esmeralda Ballesteros, and Jordi Maluquer de Motes.
Sweden: Riksbank
United States: Lawrence H. Officer and Samuel H. Williamson
United Kingdom: Gregory Clark

+++++++++++++++++++++++++++++++++++++
UPDATE (2/15/2015): I just came across this paper, which does the empirical analysis I had in mind:

Atkeson, Andrew, and Patrick J. Kehoe. 2004. "Deflation and Depression: Is There an Empirical Link?" American Economic Review, 94(2): 99-103. Ungated version here.

And here's another paper, by Jess Benhabib and Mark Spiegel, with different conclusions.
+++++++++++++++++++++++++++++++++++++

Inflation in the eurozone: goods versus services

Giulio Zanella (hat tip to John Cochrane) writes a post suggesting that Italy’s deflation is mostly imported. He supports his idea with two observations. First, the price index of goods has declined, but the price index of services has increased (coincidentally, by the same percentage). Second, among goods, those whose price has declined are: unprocessed food, energy, tobacco, and consumer durables. The first three have in common, Zanella says, that they’re traded in global commodity markets.

Assume that goods are tradable in international markets, whereas services are non-tradable—a good approximation for a country like Italy, Zanella says. Then, he concludes, the origin of Italy’s deflation is international and on the supply side, whereas the contribution of weak domestic demand is “modest.”

In this post I look at the questions of whether eurozone deflation is really “imported,” and of how big the deflation threat is.

For the currency area as a whole, disinflation is a lot faster for energy and unprocessed food than for the other components of the HICP:

Within the “core” inflation rate, durable and semi-durable non-energy goods are in outright deflation, whereas non-durable goods are not:

The identification of the tradable component of the HICP with goods, and the non-tradable component with services is not accurate, but I’ll accept it for the sake of simplicity. (After all, the main drivers of services inflation should be domestic factors.) Extending this simplification to prices, goods deflation is “imported,” whereas services deflation is “homegrown.”

The distinction is important, among other things, because the optimal monetary policy response depends on the type of deflation. If deflation is imported, the ECB should do nothing, because commodity prices and the exchange rate are outside the set of things it can (or wants to) control. Let’s look, then, at the split of the HICP between goods and services.

The sub-index for all goods was falling at an annual rate of 0.3% as of July, whereas the price index of the basket of services was increasing 1.2%:

(Goods account for about 57.2% of the eurozone-wide HICP, and services for 42.8%.) 

Services inflation appeared to be declining from 2011 up until April 2013, and since then it’s been more or less flat around 1.2% (to zoom in, adjust the dates on the chart above).

It appears, then, that a large part of the eurozone’s current disinflation might be “imported.” Another, smaller part of this disinflation is “homegrown,” but it doesn’t seem to be getting worse.

What’s the evidence across countries?

Goods inflation has declined in every single of the 15 countries I consider*. In all but two countries, prices are falling outright.

Services inflation has declined overall, but slightly (from 1.4% in July 2013 to 1.2% one year later). In five countries services inflation rose: Belgium, Ireland, France, Portugal, and Finland. In one country, Greece, deflation became less pronounced. In two countries, Germany and Austria, the services inflation rate didn’t change. And in the remaining six countries inflation declined: Estonia, Spain, Netherlands, Italy, Slovakia, and Slovenia.

So far, then, I’d say the jury is still out on whether disinflation, for services, is continuing.

Clemente De Lucia, economist at BNP Paribas, just published a research note taking a deep look at services inflation. This is what I see, from the first section of his note (“Inflation decomposition”):

1. Services inflation in core countries (Netherlands, Austria, Luxembourg, Belgium, Finland, Germany, France) is higher than in periphery countries (Greece, Spain, Portugal, Ireland, Italy). In the “core” region, services inflation shows no apparent trend since early 2011, whereas in the “periphery” services inflation has been roughly flat since late 2013:

Source: Clemente De Lucia (BNP Paribas), July-August 2014.

De Lucia mentions how, in the core, the current inflation rate is “slightly below its historical average,” and that in Germany and the Netherlands the services inflation rate is below the eurozone aggregate (although barely so):

Source: Clemente De Lucia (BNP Paribas), July-August 2014.

2. "Lowflation” is spreading across services sub-categories, but deflation is less pervasive. To gauge the diffusion of disinflation and deflation, De Lucia looks at the percentage of the 39 sub-components of the services HICP that had an inflation rate between 1% and 2%, between 0% and 1%, and below 0%.

For the eurozone as a whole, the share of services sub-categories with inflation between 1% and 2% has shot up, but the share of services in deflation has declined (from a percentage that was never higher than 20%):

Source: Clemente De Lucia (BNP Paribas), July-August 2014.

The evidence by country varies. De Lucia focuses on the total proportion of services sub-categories with inflation below 2%. That statistic is rising, De Lucia notes, in Spain, Portugal, Greece, Netherlands, and France. From that he concludes that deflationary pressure is mounting in those countries.

(Beware that, in those charts, the three categories are nested. That is, the category of "below 2% inflation" includes the category of "below 1%" inflation, and so forth. I checked.) 

De Lucia's data end in May 2014. I gathered my own data to see what's happened over the past 12 months, for the eurozone aggregate. The proportion of service categories with inflation below 2% reached a maximum in February (70%), and has declined since then. The diffusion of deflation reached 18% in December 2013, and has fallen to about 5% in July.

It will be interesting to see a breakdown by country. So far, though, I'm not convinced that "domestic deflationary pressures" are mounting.

One interpretation of the ECB’s recent policies is that the bank is not reacting to today’s decline in the price of goods. Most of that is due to past declines in commodity prices—which the ECB has no control over— and to the past rise of the exchange rate—which the ECB doesn’t target. The ECB, instead, is reacting to the disinflation of services, and the risk (or expectation?) that it might turn into outright deflation. How serious is that threat?

Based on the (admittedly backward-looking) evidence surveyed here, the risk of "homegrown" deflation is there, but it's not increasing, imminent, or pervasive across countries. (De Lucia’s note continues with a promising, long section on inflation expectations, which I haven’t had time to read yet.)

A different interpretation is that the ECB sees deflation in goods as a problem it can solve—through the exchange rate. The euro appreciated steeply, through May, and that has driven the euro price of commodities lower. The new policies might be aimed at depreciation.

But that will have to wait for another blog.

*For visual clarity, I removed from the chart the three smallest members of the eurozone: Cyprus, Malta, and Luxembourg.

Why QE might be a bad idea for the eurozone

Michael Heise, chief economist at Allianz, is against (FT) quantitative easing by the ECB. He thinks the ECB shouldn't go down the QE route, because:

First, the recent low inflation rates are in part a result of the decline in oil and other commodity prices. They also reflect necessary adjustments in the eurozone periphery. [...] There is no sign of a vicious circle of falling inflation expectations and consumer restraint. Inflation rates will gradually climb again as the economy recovers.

Second, although the ECB has several options when it comes to implementing QE, there are serious objections to all of them. Buying asset-backed securities or corporate bonds would expose the European taxpayer to credit risk.

[...]

Third, the impact of further monetary easing on output and price levels would be negligible. That is because the recession in many parts of the eurozone is caused by the hobbling effect of the unsustainable amounts of debt that were built up by public and private actors during the boom years. Over-indebted households and companies are unlikely to pile up more debt; on the contrary, they are trying to pay it down. This makes monetary policy ineffective.

Fourth, the collateral damage from ultra-loose monetary policy is accumulating. Risks to financial stability are growing as investors are piling into riskier assets in search of higher returns. Already, some assets such as junk bonds are trading at what look like inflated prices.

Fifth, further monetary easing would delay the much-needed adjustments in the balance sheets of European banks and companies. An abundance of cost-free liquidity from the central bank enables commercial lenders to continue propping up weak creditors.

Of those, I agree most with #1. A breakdown of the eurozone's consumer price index suggests that a good chunk of Europe's deflationary pressure is "imported," showing up in the price index of goods with a high component of commodities. The price index of services, on the other hand, many of which are non-tradable, doesn't indicate deflation. (More on this soon.) Yes, demand is weak, and the private sector is deleveraging, but a lot of the "deflation" problem has to do with declining prices of globally traded goods, and the appreciation of the euro between 2012 and 2014, which only started to reverse itself in May.

Of the other reasons offered by Heise, I agree with #4. I'm sympathetic with #3 (further easing will be ineffective), but I stress a different obstacle: banks are under pressure to clean up their balance sheets. In that sense, European policy is schizophrenic. On one hand, the ECB is trying to encourage more lending; on the other, regulators tell banks to improve their equity ratios.

Well worth reading

1. Housing bubble department: Switzerland, Australia.

2. Deflation, the zero-lower-bound, and multiple inflation equilibria, eurozone edition (this is a straight application of James Bullard's earlier paper, which in turn is an application of a paper by Benhabib, Schmitt-Grohé, and Uribe). Hat tip to Gavyn Davies, at the FT, for the link.

3. Puerto Rico will default (most likely), by Felix Salmon.

4. U.S. labor market spider chart. Great visual tool by the Atlanta Fed. Don't miss the jobs calculator, and the inflation dashboard!

5. U.S. small businesses don't create jobs any more, from BloombergBusinessweek.