Falling population shares: a highly-productive big city

Writing about Wales the other day I included this chart

wales 1

The comparable chart for Scotland is even more stark (16 per cent of the Great Britain population in 1801 and just over 8 per cent now).

But what really caught my eye when pulling together the numbers was this chart.

london 19.png

I guess part of my brain knew that greater London’s population had fallen for several decades, but that bit never quite connected with the bits thinking about world cities, agglomeration and so on and so forth.  London is one of the great world cities, a key financial centre in an age when capital is more mobile than it was for decades after the war.  There is no other really great city in the UK, the UK’s population hasn’t increased that rapidly by New World standards, and yet the share of the UK population resident in greater London is less now than it was for decades prior to World War Two (true even using the orange dot –  for which there is no time series – the estimate of the population of the (defined by contiguity of population rather than local authority boundaries) of the greater London urban area.

(As it happens, on checking one finds that the New York metropolitan area population is also lower now, as a percentage of the total US population, than it was several decades ago – I could only see data back to 1950.  But the US is different  –  there are multiple very large cities and the spread of air-conditioning greatly affected the liveability of many of those places.)

As you may recall from Saturday’s post, estimated GDP per capita in London is 188 per cent of that of the EU as a whole (and about 180 per cent of the UK as a whole).  The only other (Eurostat-defined) region that comes even close to London is (close to London) “Berkshire, Buckinghamshire, and Oxfordshire” (at 151 per cent of EU as a whole).

These have the feel of places where if more people were able to live there more people would be better off.  The whole of the UK might even be better off on average (a larger proportion of the population able to do more highly-productive jobs), even if the London premium over the rest of the country narrowed somewhat.

And yet, of course, as everyone knows London house prices are really expensive –  price to income ratios similar to those in Auckland (with incomes higher), typically for small houses and small sections.  You can tell similar stories about San Francisco/San Jose or New York (where GDP per capita are well above those of the US as a whole).   Rigged housing and land markets really seem to have visible consequences in pricing people out of working in highly productive cities.

Where the story is much less compelling is in Auckland (or Sydney or Melbourne). I wish it were otherwise –  I’m a strong supporter of land use liberalisation –  but

(a) on the one hand, the populations of those cities (urban areas) have actually increased very substantially as a share of national population (especially Auckland: 8.5 per cent of the population in 1901, and about 33.5 per cent now), and

(b) in none of the Australasian cities do the estimates for GDP per capita show up with any very substantial margin over the rest of the country (see, by contrast, London above).  People who just don’t earn that much (or produce that much) have found a way to live in those cities anyway.

Fixing the New Zealand urban housing markets is, or should be, a matter of dealing to one of the grosser injustices in our economic system, but it is far from obvious that there is a compelling case in issues around productivity and wider economic performance.  If anything, there are probably already more people in Auckland – and perhaps Sydney/Melbourne –  that there really are highly-productive opportunities that are either waiting for them now or would spring up were housing once again as affordable as it should be.

 

Advanced countries with flat/falling populations seem to do just fine

In my post on Monday I was critical of various aspects of Liam Dann’s Herald pieces in praise of New Zealand’s high rates of immigration.  Part of his story was that we simply had to keep on with high rates of immigration or our population would stop growing and……well, there lies dragons, or at very least “economic stagnation” and some existential threat to “New Zealand’s economic and social wellbeing”.

A casual reader might have supposed that there were no examples of countries with flat or falling populations, no straws in the wind we could look at and see how likely it was that a stable or even modestly falling population would represent a serious threat to the living standards (material and otherwise) of New Zealanders.

There was a new wave of Conference Board Total Economy Database data out a couple of weeks ago.  It has wider coverage than the OECD databases and the economic estimates are a bit more timely too.  I’ve used Conference Board data in numerous posts over the years, with a particular focus on the 40 or so advanced countries (OECD members, EU members, plus Singapore and Taiwan).

Over the last 20 years, 10 of those countries have experienced a fall in the total population, and another two have had almost no population growth.

population falls.png

If one takes a more recent period –  just the current decade –  all the countries with falling populations are still falling, and they’ve been joined by Portugal. Spain’s population is also now flat.

Of course, the economic performance of one of these countries –  Greece –  has been truly atrocious.  Real GDP per capita is still about 20 per cent below 2007 levels, and even the average level of labour productivity has fallen.  But no one supposes that Greece’s economic woes are because the population is flat or falling: if anything plummeting living standards and high unemployment have prompted Greeks to look for better opportunities elsewhere.

Here is the productivity growth (real GDP per hour worked) performance of those countries with flat or falling populations, again over the 20 years to 2018.

population 2.png

A flat or falling population is, of course, no guarantee of economic success (Greece and Portugal are what they are), but it certainly doesn’t seem to have been a major roadblock in the way of strong economic performance over the last 20 years.   Even Japan –  already rich 20 years ago, and often a poster-child for the alleged economic problems of a falling population – has had productivity growth outstripping that of the median advanced country.

Where would  New Zealand fit in that picture?  New Zealand –  with rapid population growth – managed 24 per cent productivity growth over 20 years, better than Greece and Portugal, but well below the median, let alone the median of the flat/falling population countries.

20 years ago falling populations really were a new phenomenon.  And 20 years ago, many of those countries really were rather poor, just a few years out of Soviet domination.  Perhaps one needs to look at more recent periods to really see the (alleged) crippling effects of a flat or falling population?  So I had a look at the period from 2007 to 2018 (choosing 2007 so as not to start my comparison in the middle of a severe recession), and over that period the median country with a flat or falling population also did materially better than the median advanced country (or the median of the countries with fast population growth).    New Zealand, once again, underperformed each of those medians.

But the focus of Liam Dann’s article had been on the population/immigration surge New Zealand has experienced since 2012.    I’m very reluctant to put much weight on short-term comparisons (even across a pool of other countries there can be other cyclical factors that muddy the water), but….what the heck, here it is.

You’ll recall my chart showing an estimate of labour productivity growth in New Zealand.

GDP phw may 19

That was basically no productivity growth over the last five years, and perhaps 1 per cent total productivity growth over the period since 2012.

There are various ways of getting an estimate of labour productivity. Mine (in the chart above) averages the two measures of GDP (production and expenditure) and the two hours measures (HLFS and QES).  I’m not sure quite what the Conference Board uses, but their numbers aren’t inconsistent (if perhaps a touch lower) than what is in my chart.

Here is productivity growth for the countries with flat and falling populations from 2012 to 2018, with numbers for New Zealand, all advanced countries, and the median of the flat/falling population countries also shown.

popn 3

Using slightly different estimates, we might have done better than Portugal but that is as far up the ranking as you can get New Zealand over this period, one when we have –  on the Dann telling –  been blessed by such a beneficient immigration policy (and associated rapid population growth).  Of these countries, Japan and Slovakia already have higher average labour productivity than New Zealand does, while many of the countries are now also close.   The convergence story in defence of New Zealand (others are just catching up) has long since lost most of its salience: we were supposed to be one of the countries catching up, but we just haven’t been.

For what it is worth, over this particular recent six year window, New Zealand’s productivity growth was not just second lowest on this chart, but second lowest among all the advanced economies.

My main interest is in New Zealand, an incredibly remote set of islands. Over decades now there has been no sign that rapid policy-driven population growth has been helpful to our medium-term economic performance.  But there is no necessary reason why issues that might be relevant to our economic underperformance should also be relevant for countries much closer to major markets, supply chains, networks and opportunities.

On the other hand, there is no sign that countries with flat or falling populations are doing particularly poorly.  In fact, in economic terms, most seem to have been doing just fine.

Simple cross-country correlations can always only take one so far.  After all, the countries with flat or falling populations will include those where people are fleeing underperformance (Greece say) and countries with rising populations will include some of those where people are attracted to economic success (Singapore say): in neither case is it likely that the main direction of causation runs from population growth to economic success.

But, for what they are worth, here is a scatter plot showing population growth and productivity growth across those 40 or so advanced countries over 1998 to 2018 (one dot per country, New Zealand is red).

popn 4

It isn’t a tight relationship, but it is there (and was there is the economics literature decades ago) and isn’t obviously skewed by a single outlier country.  And New Zealand isn’t an outlier either – our productivity growth over 20 years was only a bit less than one might have expected from this crude relationship.

For the much shorter more-recent period (2012 to 2018), the negative relationship is still there but, as one would expect (with other stuff going on), is weaker.    But New Zealand more starkly underperforms.    Perhaps that underperformance –  little or no productivity growth for years –  will eventually be revised away.  Perhaps.

I’m not one of those with any generalised aversion to population growth.  Most population alarmism, at least at the macro level, is misplaced.  Technology, ideas etc keep on allowing for rising material living standards for more people.  But equally, there is little evidence that rising populations –  beyond some critical low thresholds –   themselves work to boost material living standards, and some signs that advanced countries with rapid population growth do less well (in material terms) than countries with less rapid population growth (even with all the sometimes conflicting chains of causation at work).

But across advanced countries as a whole even if all that was simply false, we’d still be left with a picture of New Zealand where policy has fuelled rapid population growth for most of the last 70 years, even as our relative economic performance has kept on declining.   Whatever the situation in Japan or Slovakia, there is decent prima facie reason to be intensely sceptical of the alleged economic gains to New Zealanders from continued high policy-induced immigration to this extremely remote corner of the world.

And few/no signs that countries with flat or even falling populations need to worry about economic underperformance stemming from such population changes.

Population size and GDP per capita: US states

There have been a few posts here (here, here, and here) in the last week or so around the issue of population size and GDP per capita –  not because my prior is that there is any such relationship but (a) because I think there isn’t, and it is worth occasionally illustrating that across countries, and (b) because even some officials in the New Zealand government still appear to believe that our (small) size is a material element in the story of what holds New Zealand back.   There are arguments why, in theory, a larger population might result in better long-term economic performance (higher productivity), but whatever the merits of those arguments they seem to have been outweighed by other factors.  In the world we currently live in, the average big country is no more economically successful than the average small country, and that is so whether one throws all 190 countries into the mix at once, or looks only at advanced countries, only at European countries,  only at moderately-sized countries (eg excluding China and India) or whatever.  And one might imagine a plausible story in which economic success might have led to more population (migration or natural increase) rather than the other way round.

But what does the picture look like for US states?  Across all 50 states (one dot per state) here is the cross-sectional picture.

US states 2

And again, no obvious or statistically significant relationship.   In fact, to make the dots a bit easier to see, I left off the (non-state) District of Columbia which with about 600000 people has per capita GDP of about $160000.

The US case is interesting because –  unlike the situation with comparisons across individual countries –  there is no legal obstacle in the way of US citizens (and residents) moving within the country (across state lines) to pursue opportunities.   In that sense one might have (actually I initially did) expected to see more populous states also being more economically successful, if only because of internal migration.  But the relationship between population size and GDP per capita seems about as weak across US states as it does across countries.

I wouldn’t want to make anything much of these US state comparisons –  I just did them because the data were there, and having dug it out I thought I’d share it.  After all, in various places, big metropolitan areas straddle two or more states (New York is the best known example).   Then again, there is more policy similarity across US states than there is even across member states of the EU –  so a few more things are help constant when one looks at the simple scatter plot this time.

Whatever the theoretical arguments, a bit of history suggests we shouldn’t be surprised at the lack of any sustained relationship between population and economic performance.  After all, at its 14th century peak –  as probably the richest place in Europe – Florence is estimated to have had a population of fewer than 100000 people.

 

Small size simply isn’t the issue

Just yesterday I wrote, in response to a comment, that

My point simply was that there is no obvious correlation, in the cross section, between population size and GDP per capita (or productivity). I’m not aware of any serious observer arguing otherwise

At the level of very simple correlations, I’d illustrated this lack of relationship –  whether for all countries, just advanced countries, excluding the handful of extremely large countries, or whatever.  Bigger countries (by population) don’t, on average, have higher per capita incomes than smaller countries.

But then I went to a seminar at the Productivity Commission yesterday afternoon, attended by various private and public sector people.  The substance of the seminar –  a new MBIE report on the manufacturing sector, and the discussion of it and of possible policy responses –  is embargoed until David Parker releases the report next week.   But I hope I’m not breaching any rules in simply reporting that I noted around the room an almost unquestioned acceptance that size (small) is one of New Zealand’s economic problems.  Normally I might barely have noticed it, but having been writing about the topic in the last few days, it niggled away at me.

The more I’ve thought about that issue over the years, the more I’ve concluded that those who hold it are simply wrong, and perhaps in the thrall of the political equivalent of Keynes’s “defunct economists”  –  the long tradition of political leaders, dating back at least to Vogel, who’ve wanted, as a matter of policy, a lot more people in New Zealand, believing (presumably) that New Zealanders would be better off as a result.  There doesn’t seem to be much –  any? –  evidence in support of any economywide economic benefits flowing from this preference.

One hears talk in such discussions of ideas like “markets work better in big economies”, or even talk of economies of scale –  or opportunities for specialisation – in government/regulation.  In principle, the arguments sound plausible enough.

But they rarely seem to confront the simplest stylised data.  For example, Australia and New Zealand are almost equally remote (on standard measures), but Australia’s population has been consistently much higher than ours, and yet for a century (say 1860 to 1960) material living standards (GDP per capita) were much the same on the two sides of the Tasman.

Or we could look at some advanced countries where distance/remoteness is much less of an issue.  In what follows I’ve looked at the OECD member countries in Europe (continental plus the UK and Ireland).  There are 24 of them, ranging in population from Germany’s 82 million to Luxembourg’s 0.6 million.  Five of these countries – New Zealand’s size or smaller – didn’t even exist as independent states thirty years ago.

In this chart –  for 2016 –  I’ve ordered those 24 countries by population size and shown the real GDP per capita for each.  I’ve also shown the median (so not distorted by Luxembourg, or issues around Irish tax) for the group of countries with a population less than 10 million, and also the median for the large European OECD countries.

europe real GDP and popn

If anything, the typical larger country has a lower per capita income than the typical smaller country.  Of course, these are small samples, so not much weight should be put on them, but there is nothing to suggest bigger countries are performing better than smaller countries.  And if one insisted on excluding the former communist countries – even though they are now 25 years on as market economies –  the gap (in favour of the smaller countries) – is larger.

In many respects,  real GDP per hour worked  (labour productivity) is a better metric of economic performance.  Here is the same chart, for 2016, using the OECD’s data on productivity.

europe real gdp phw and popn

I could exclude Ireland and Luxembourg, I could exclude the ex-communist countries, I could add to the “small” category the countries (Portugal to Belgium) with populations just over 10 million, and it won’t change the story.  There is nothing in the simple stylised facts of European OECD countries suggesting that bigger countries do better than smaller ones.  Even at the most prosperous core bit of Europe (London is the richest –  by income –  region in Europe), Belgium, Switzerland, Denmark and Austria do really well.  And so do France and Germany (the UK less so).

Of course, there are lots of other things that help explain any individual country’s performance. Norway, for example, wouldn’t rank so high without oil, or Ireland without the features of its corporate tax policy which see a lot more economic activity booked in Ireland than directly results from economic activity occurring in Ireland.  One issue, of course, is the quality of policy.  There are lots of different dimensions of that, and sometimes one sees a story in which small countries try harder, regulate less, or whatever to overcome the alleged disadvantages of size.   One widely-used indicator is the OECD’s index of product market regulation.  As it happens, the PMR score for the median small country in OECD Europe is less good (ie more less-liberal regulation exists) than in the median large country.

Smarter people with richer datasets and serious econometric skills can produce much more complex models, encapsulating a lot more information simultaneously.  But whatever the results of such models –  which often end up depending on the modellers’ embedded assumptions –  it is always worth bringing them back to check against the simplest stylised facts.  Even in a region where distance is much less of a differentiator among countries (it isn’t nothing –  Portugal and Greece would have it tougher than Belgium and the Netherlands even with great policy) population size doesn’t seem to be an advantage, and isn’t associated with either higher GDP per capita or higher productivity.

For decades, I’ve used the line that if only we could detach New Zealand from the ocean floor and relocate it –  land and all –  in the Bay of Biscay, just off the coast of France we’d be much better off materially, all else unchanged.  Perhaps the North Sea would be even more propitious.  But the point remains, the biggest handicap to economic success in New Zealand is our distance/physical remoteness –  in an age when, across the board (although with individual pockets otherwise) distance isn’t becoming less of an issue, but perhaps even more of one.    A modest number of people probably can do very well here, but not many.   And yet our policymakers –  aided and abetted by official advisers –  keep driving policy to locate ever more people in a really quite unpropitious –  even if beautiful and (now) peaceful –  location.

There is simply no evidence supporting the notion that our small size is “the problem” (or even a material part of it), and when the story continues to be invoked it simply serves as a distraction –  mostly unwittingly so –  from a focus on the real issue, responding realistically to the unchangeable (absent quite different technologies) constraints of our physical isolation.

More on population and per capita GDP

My quick post on Saturday, in response to someone’s comment, was designed simply to illustrate what should have been quite an obvious point: looking across countries in any particular year, countries with large populations don’t tend to be richer (per capita GDP) than countries with small populations.  Just among the very big countries, the United States is towards the top of the GDP per capita rankings (beaten by a bunch of small countries), and China, India, Indonesia, Pakistan, and Brazil are not.   Since both the physical sizes of countries, and their populations, are the outcomes of all sorts of historical factors, it wasn’t an observation about immigration policy or (wince) “population policy”.    And, of course, GDP per capita isn’t everything: moderately well-off large countries are typically more powerful (defence or offence) than small rich ones.

But, continuing to play with the same data download from the IMF WEO database (190 or so countries and territories) for the period since 1995, is there any obvious relationship betweeen population growth rates, and growth in real per capita GDP?

Here is the chart

population 9 Aug 1 all countries

Actually, I’ve left out three extreme outliers –  all oil producers.  Equatorial Guinea had growth in real GDP per capita of about 2500 per cent over the period, and UAE and Qatar had population growth of 300-400 per cent (in one case, with falling real GDP per capita, and in the other with a moderate increase).

There is basically no relationship between the two series –  again, each dot is a country.  The simple linear regression line is downward-sloping but that probably wouldn’t be a statistically significant relationship.  Bear in mind though that simply charting population growth and per capita GDP growth for the same period could have shown an upward-sloping relationship even if there was no causal link from faster population growth to faster per capita economic growth: countries with rapid growth in real GDP per capita might be expected to attract more people (immigrants flow towards opportunity) and perhaps even induce higher birth rates.   But there is no sign of even that sort of relationship, across all countries, and over this period of 20 years or so (this sort of reverse causality is a big problem looking at annual data, but much less so looking at long periods).

The full sample of countries includes a huge range of types of countries –  from war-torn poverty stricken basket cases (Syria, Somalia, Afghanistan) to tiny remote islands (Tuvalu), as well as places with strong institutions, good connections, and an established record of economic performance.   How do this simple bivariate relationships look if we focus just on these latter countries?

In various posts over the years, I’ve used a sample of about 40 fairly advanced countries, encompassing the members of the OECD and the EU, as well as Taiwan and Singapore.   There is still a lot of difference among these countries: places that were non-market communist economies only 30 years ago, the odd place (eg Mexico) that is more like an honorary member of the group of advanced countries, as well as the places with very high productivity (France, Germany, United States, Ireland, Norway).  And there are countries as small as Malta or Luxembourg, and as large as the United States.  This group leaves out countries that appear to be rich only because of oil.

Here is the simplest plot: of the levels of population and GDP per capita in 2016.

popn advanced 1

This time, the simple regression line is very slightly upward sloping.  Remove the US and it changes sign.  Remove all the countries with more than 50 million population and it is still downward sloping.

But what about growth rates?  For quite a few of these countries, GDP per capita is pretty shaky before about 1995 (communist-era and immediate post-communist transition).  That’s why I’ve done all these charts for just the last 20 years or so.  But that also happens to be a period when there has been a lot more population movement between advanced countries (especially in Europe).

popn advanced 2

Recall that even if there was no causal relationship running from population growth to growth in real GDP per capita, there was a possibility that we might have seen an upward-sloping relationship simply from any link between successful economies drawing in more people (as happened in Ireland most obviously –  net immigration rising well after the boom in per capita GDP and productivity).  But, in fact, across these 40 or so advanced countries, any relationship is downward sloping.  Across these countries in this period, faster population growth has been associated with slower real per capita GDP growth. (For the eagle-eyed among you, New Zealand is the red dot.)

And in this final chart, I’ve broken the period in two.  I’ve charted, for each country, population growth in the first 10 years of the period (1995 to 2005) against real GDP per capita in the second half of the period (2005 to 2016).  In other words, none of the population growth variable is directly caused by the growth in the real per capita GDP variable.

popn advanced 3

The downward-sloping relationships is weaker this time –  less of the variance in GDP growth is explained simply by prior population growth –  but again it is downward-sloping, and not the upward-sloping line many of the immigration-policy boosters in New Zealand would like us to believe.   I forgot to mark New Zealand on this chart –  we are one of the dots a bit below the line with 12 per cent population growth –  but there is nothing unusual about New Zealand’s place on either chart.

If one wants to get more sophisticated, one could look at growth in labour productivity or total factor productivity, rather than just real GDP per capita.  Especially for TFP, one becomes dependent on the model used in estimating TFP and the available sample of countries drops away,  As for labour productivity, in earlier posts I’ve illustrated the lack of a positive relationship between population growth and productivity growth (and recall that New Zealand has managed better real per capita GDP growth than productivity growth, by working longer hours), and that –  for example –  if anything business investment as a share of GDP has been negatively correlated with population growth across advanced countries.  That is the opposite of what might been expected if population growth –  and immigration –  was typically boosting the productivity and per capita income prospects of recipient countries.

It is past time we started backing our own people, not looking to replace or augment them with a mythical group from across the water.  Part of that involves the government getting serious about facing up to the disappointing economic outcomes of our long-running Think Big economic and social experiment with large scale immigration.  As part of that, in turn, a serious review of immigration policy by the Productivity Commission –  there have been two in Australia in the last 15 years –  would be a good place to start.

 

Population and real GDP per capita

I noticed a few comments to another of my posts about possible links between population size and economic performance.  My working assumption is that, on average, across all countries, there isn’t any such relationship.   Apart from anything else, if there were a positive relationship –  that was more than chance –  it would suggest that two countries merging would increase their respective real incomes.  And yet for at least the last 70 years, we’ve had steadily more countries emerging.  No doubt economics isn’t the only thing at work in those choices –  people might be willing to pay a price to be “free” and self-governing –  but it isn’t likely to be an irrelevant consideration either.

But what do the data show?   Here I’ve just used the IMF World Economic Outlook database data for 2016.

The first chart shows the relationship –  for the 193 countries/territories the IMF reports data for –  between real GDP per capita (in purchasing power parity terms) and population (each dot is one country).   The population term is expressed in logs.

popn and real GDP pc

As (I would have) expected, there is basically no relationship at all.   The simple linear regression line is actually slightly downward sloping, but that won’t pass any test of statistical significance.  Perhaps one could craft a story in which the top 10 countries (in terms of per capita income) all have quite small populations –  the biggest is around 5 million people – but since oil plays a big part in most of those individual cases even then one shouldn’t make too much of the point.

And here is the chart if we look only at the countries with populations from 0.5 million (a tenth of New Zealand’s) to 50 million (ten times New Zealand’s).  Since that is a much more compressed scale –  not everything from Tuvalu to the People’s Republic of China –  this time the population variable isn’t expressed in logs.

popn and real GDP pc 0.5 to 50m For those with sharp eyesight, New Zealand is a dot coloured orange.

Again, there really isn’t any sort of relationship.  Again, the simple regression line is downward sloping, but there are lots of countries with very small populations and very low per capita incomes.   But even within this more-compressed range of populations, there is no sign at all of any sort of upward sloping relationship –  the idea that, on average, a higher population will be associated with higher per capita incomes.

Of course, within each of these dots there are complex historical relationships, as to how population in any particular country came to be what it was (some about conquest, making big countries out of small one; sometimes the historical carrying capacity of the land; in some the role of slavery (eg forced depopulation from Africa), in others the role of immigration policy.   Some locations offer better prospects than others and will, typically, have attracted or retained poeple accordingly.

But this post isn’t attempting to get into any of that. it is simply observing that at the most elementary level of numerical analysis there is no sign that countries with larger populations tend to be richer (whether as a matter of cause, or of effect).

(Not much) investment in New Zealand

A few days ago I ran a post on the cross-country relationships between population growth on the one hand, and residential, government, and business investment on the other.   Using OECD data, averaged for each country over a couple of decades, it was apparent that (a) as one would expect, residential investment makes up a larger share of GDP in countries with faster population growth (people want a roof over their head, but (b) business investment as a share of GDP was smaller the faster the population growth a country had experienced.   New Zealand’s experience was quite consistent with these relationships.  That should prompt some introspection on the part of those –  bureaucrats, politicians, and other lobby groups –  who champion our large-scale non-citizen immigration programme, the largest such active migration programme (at least for economic reasons) in per capita terms anywhere in the world.

But today, I justed wanted to look at New Zealand’s own data on investment, and particularly the experience in the current cycle.    My starting point is this chart, using the components of gross fixed capital formation (“fixed investment” in the national accounts), as a share of GDP, going back to the 1987 when the official quarterly national accounts begin.

GFCF components to Mar 17

As I noted the other day, “business investment” isn’t an official SNZ category –  it would be great if they actually started publishing one –  but instead follows the OECD practice of subtracting general government investment (schools, roads etc) and residential investment from total investment.     It isn’t fully accurate, to the extent that some residential investment is done directly for the government (so there is some double-counting) but (a) the effect should be small, and (b) it is a consistent treatment through time.

And in case anyone is wondering what the spikes in 1997 and 1999 are, they are navy frigates.

Three things struck me from this chart.

  • First, total investment as a share of GDP (the grey line) has been rising quite strongly from the trough in 2009 and 2010, but
  • Second, total investment ex residential investment (the orange line) has barely recovered at all, and
  • Third, business investment (as proxied by the blue line) has not only barely recovered, but is now smaller as a share of GDP than in every single quarter from 1992 to 2008.   And this even though our population growth rate has accelerated strongly, to the fastest rate experienced since the early 1970s

The difference between the orange and grey line is residential investment.   It has picked up a lot as a share of GDP, but then it would have been extremely worrying if that were not the case.  After all, we had a series of destructive earthquakes in Canterbury, and huge volume of resources had to be devoted to simply restoring the existing housing stock.  And we’ve had a big acceleration in population growth.    Residential investment as a share of GDP is now higher than at any time in thirty years, although house and land price developments suggest that residential land is still being held artificially scarce.

Businesses invest when they see opportunities and can raise the finance (internally or externally to take advantage of the opportunities).     There will always be some financing constraints –  firms that don’t have the retained earnings or can’t persuade someone else to provide additional debt or equity –  but it is a little hard to believe that, as this stage of the cycle, those financing constraints are much different than usual.  It suggests that firms just don’t see the investment opportunities in New Zealand to anything like the extent they once did, even though the population is growing as fast as it ever has in modern times.     It is at least suggestive that the persistently high real exchange rate might be an important part of the explanation.

New Zealand’s quarterly national accounts data go back only to 1987, but the annual national accounts data go back to the year to March 1972.    Here is business investment as a share of GDP right up to the year to March 2017.

business investment to mar 17

Not much above recessionary levels (1991 or 2009), and showing no sign whatever of picking up.   And that is even though the population (and employment) are now much higher than would have been foreseen just a few years ago.    Investment goods do appear to have got (relatively) cheaper over time, but that seems unlikely to adequately explain how firms saw investment opportunities of around 12 per cent of GDP in the two growth phases, but only around 10 per cent now  (especially as we know we’ve now had no productivity growth for five years).

Statistics New Zealand also produces annual estimates of the capital stock.  The latest observation is for the end of March 2016, but the earlier charts suggest there is little reason to think the story for the most recent year will be any more encouraging when the March 2017 data are released later this year.  This chart shows the annual growth rate is the estimated per capita real net capital stock (excluding residential dwellings).

cap stock growth

This indicator uses all the non-residential capital stock (ie including that belonging to the government sector).  As government investment has held up more strongly than business investment (see the first chart above) and as employment has been rising faster than population, the picture for business investment per employee would probably look even more disconcerting.

And, of course, all the official capital stock numbers use reproducible capital only.  In New Zealand, in particular, land is a major input to significant parts of business production.   The quantity of land is fixed (improvements to the land are included in the investment numbers above), and that fixed quantity is spread over ever more people.

Given our very serious housing situation, with house price to income ratios among the highest anywhere in the advanced world, it should be a bit troubling when really the least poor bit in the investment data is residential investment.   But lest I inadvertently comes across sounding upbeat on that score, here is annual growth in the SNZ real residential capital stock per capita.

res cap stcok

But perhaps this too is some sort of “sign of sucess” or “quality problem”?    Most people, I suspect, would settle for signs that if we are going to have rapid policy-driven population growth, that businesses would then find it remunerative to invest much more heavily, whether in building houses or producing other stuff to sell here or abroad.