Bad reasons to delay a stimulus package
Over at Worthwhile Canadian Initiaitive, Stephen and Nick have been making the point that “Canada is not the United States, we are not in recession (although the odds that we will be soon are much better than even), and there’s no reason to rush into a program of public works”.
To some extent, they are right, and that go-slow approach would seem at the heart of the Conservatives’ economic plans. But it is a cute technical point that in proper context is wrong.
It does take some time to get infrastructure projects up and running. It requires cooperation with provincial governments and municipalities, and time for planning processes to get underway. That is the knock against infrastructure projects in normal recessions. But our times are not normal. There is every reason to believe that 2009 is going to get ugly, and so we need a signal to other levels of government to get going. Every other country in the world has recognized this, and is acting accordingly. Canada needs to play its part not free ride off the effort of other countries.
The rest is a matter of reinforcing the automatic stabilizers, primarily EI, so that as the downturn happens they kick in. The feds should immediately make changes that open up accessibility, increase benefits and lengthen the time for which they can be claimed.
These things should not wait until late-January 2009.
indeed.
not the time to fiddle while things burn.
But things aren’t burning.
And how can we be certain that when things go up in flames that it will be the construction sector that will be first and hardest hit?
I forgot to mention the line that Le Soleil columnist Gilbert Lavoie used awhile ago (my translation):
let’s wait for the storm to hit before deciding whether we need to buy umbrellas or snowsuits
The increase in residential construction investment in recent years relative to historical baselines is about 2 percentage points of GDP (slightly more than that at the peak of the boom, as it has already started to come down). This crisis is rooted in a housing bubble. That bubble is bursting, albeit more slowly than in the US and probably less inflated to begin with.
So infrastructure/construction is a good place to start. It is also accessible to workers laid off in other blue-collar industries, like autos, steel, etc.
My question back to Stephen is why sit on our hands when we have a pretty good sense of what’s to come. There is bigger downside risk in doing nothing than there is in preparing and, if we are lucky, over-reacting. If we know the storm is coming why wait until we get wet and have to rush out to buy an umbrella? Or perhaps more aptly, why wait until your home starts leaking before fixing the roof?
But we *don’t* what’s to come. The construction sector is already working flat out; what’s the point of floating new infrastructure projects that no-one has the time to build? How do you know that significant slack will show up there in the next two months? How do you know that unemployment in the construction sector will be the most pressing issue in the next two months? More than (say) the auto sector?
I don’t think you do. I don’t think anyone can at this point. So why commit now when we don’t have to?
And it’s not fair to say that I’m advocating doing nothing. I’m advocating doing nothing we’d regret.
Like I said, it takes time to get these things up and running, so it is prudent to start planning now, especially given the pace of federal-provincial negotiations.
I don’t think we would regret retrofitting our infrastructure for a more climate-friendly age, even if that were to crowd out some private sector investment.
In terms of “knowing” I can point to data on the slowdown, but anecdotally I just had a hot water heater installed and those guys are experiencing a major drop off in work as people hold back on making new purchases. We can also look at what’s happening in the US, the UK and other countries who are ahead of us but facing the same dynamic in terms of housing markets.
But Canada is not the US or the UK.
Got any data suggesting significant slack in the construction sector? Or that significant slack will develop within two months? Or that when the recession comes, it will hit hardest in the construction sector? If not, then why insist now on help for that one sector?
I think the problem is a lack of imagination. The old playbook that we’ve been reading for generations use public works as the standard mechanism for fiscal policy, because it was easy to understand, and the story came complete with a picture of the Hoover Dam.
This is 2008, and the construction sector is working at full capacity. Time to write a new play or two.
Canada is not the US or the UK but the underlying economic problem is the same. Looking in the rearview mirror is not that helpful when headed for a cliff.
Ultimately a stimulus package is about much more than infrastructure and pinning it all on construction employment seems disingenuous to me. I outlined six areas for consideration in my recent economic and fiscal update for the CCPA, of which infrastructure was one.
Anyway, in terms of data, here’s what I wrote as it pertains to that part:
Economic weakness is particularly evident in declining in-
vestment. Investment in residential construction has dropped
in both first and second quarters of 2008, signaling an end to
the housing boom of recent years. At the end of the 1990s,
investment in residential structures was about 4.6% of GDP.
That figure swelled to 7.1% in 2007, and has slipped back to
6.8% as of the second quarter of 2008. A continued decline
back to normal levels will have major implications for Canada’s
overall economic performance, with essentially two percent-
age points of GDP at stake. A contraction of this scale would
have major implications for a labour market that has boomed
with construction jobs in recent years.
While the Canadian housing situation is not nearly as grim
as what has been unfolding in the U.S., the Canadian market
has clearly passed a turning point. The supply of homes for
sale has grown in major Canadian cities, while demand has declined, leading to drops in sales and in resale values for the first time in many years. Recent data point to a 6% drop in average
resale prices in September compared to a year earlier. Price
declines are likely to continue through 2009, if not longer. The
spillover of falling home prices will be to reduce employment
in new residential construction (which had already declined in
the first half of 2008).
Stephen Gordon, you do realize that we need massive expenditure for low cost public housing, and public water infrastructure in almost all Canadian communities, particularly First Nations’ communities.
You might also want to take at look at work of the Earth Policy Institute on the potential of green energy infrastructure.
You also realize that there are hundreds of thousands of unemployed people who have been laid off from various sectors, who can be trained and employed in needed infrastructure.
Actually, many people are burning, in frustration at the stupidity and complacency of some in higher office. The earth too could use a bit of chill.
I could use stronger language, but the diddling of some economists is less tha useful. Why must it always be the grassroots which has to do the heavy lifting?
We’re calling on economists of progressive stripe to get moving, pick up steam, get digging, produce reports, so we don’t have to fight you as well as Harper to get some decent supports.
Ultimately a stimulus package is about much more than infrastructure
It’s only the infrastructure part – in particular, the insistence that its form and size must be decided now, and that we can’t afford to take two months to think about what, where and how much – that bothers me. I’ve been pretty clear about supporting the parts about strengthening EI and beefing up transfer programs (Child Tax Credit, GST rebate, Working Income Tax Benefit).
In case you’re wondering why the notion that we must put together an infrastructure program in great haste bothers me, here’s why: it starts a game in which those with the sharpest elbows and the best political connections win.
That’s not the sort of game in which the marginalised parts of society does well.
I wonder how outside of beefing up UI and getting a coordinated training and labour planning regime off the ground the government can do much in the auto and forestry sectors? Both sectors are plagued by long term structural problems on both th demand and supply sides of the equation who’s ultimate resolution will come long after the crisis and recession has passed. Moreover, construction has just started to contract so it will be a while before noticeable labour surpluses build-up there.
So perhaps SG does have a point: We probably need more contingency type planning than a pre-determined stimulus packages at this stage. That is direct transfers to low income and unemployed Canadians is perhaps the sole thing, in the immediate term the state can deliver with any efficiency.
That said, I would go further than SG and argue that we need to start planning for serious industrial planning. First, because as a I argued above the present problems in autos and forestry go way beyond a short term stimulus package. And second because it is not all clear that at present the CDN state has the planning or delivery capacity to carry out serious industrial stewardship at this time. 30 years of planning retrenchment from what was by all observes an already weak state in the planning area does not bode well for a stimulus package which is being touted as a structural fix for the future.
I have this weird vision of Dion and Layton on an Aircraft Carrier claiming “stimulus accomplished” after the first reading of their package.
Just to be clear, I DO think we need a stimulus package, and right now, and a big one.
But I want a monetary stimulus package, not a fiscal stimulus package. We can still cut the overnight rate from its current 2.25% to 0% if needed (plus quantitative easing on top of that).
When we’ve used up all the monetary ammunition (and that might happen), then we will need a fiscal stimulus. The particular type of fiscal stimulus can be decided more on micro grounds: EI, GST, infrastructure (if that’s the best use of $, and if construction employment collapses).
Nick
What is the evidence that monetary stimulus is working?
Travis:
I would love to talk your ears off for hours on that subject. I have some very weird views on that question! Here’s the very short version:
In the normal sense of evidence: there is no evidence whatsoever that monetary policy is working, and there never could be any evidence that monetary policy ever works, during a regime of inflation targeting. (Told you I had weird views, didn’t I!)
Think of an experimental design that a psychologist would set up to properly test whether MP works. Ideally, you hold all the other things constant while you vary the control variable, and failing that, you randomise the control variable so it is uncorrelated with all the other variables which might affect the results. But that is exactly the opposite of what the Bank of Canada does under inflation targeting. The last 16 years of data on monetary policy, inflation and output, are like the result of a psychologists’ April Fool’s day joke: where they tried to think up the worst possible experimental design to examine the effects of monetary policy.
The only evidence I do have that monetary policy works is this. If I see a car being driven down a winding road, and it does not crash, I assume that the steering works, and that the driver knows which way to turn the wheel if he wants to steer right or left. Same with monetary policy and the Bank of Canada.
OK, there’s other evidence as well, from other countries and other times, when you put an idiot in charge of the central bank, so that you do get something closer to the psychologists’ experiment.
Sorry. I told you I had weird views on that question!
By the way, if interest rates do hit the zero lower bound, then I predict that monetary policy won’t work any further, and the car will crash, unless we use fiscal policy as well (or get very lucky).
And that was the short (and overly simplified) version. There’s a Bank of Canada working Paper, 2002 I think, with a bit longer version. And a much longer version on my computer.
Nick,
I do not doubt that MP works. Could I fairly caption your position on the evidence for MP as the “Rowe Revealed Effectiveness of Monetary Policy theorem” or RREMP?
My question was what evidence do we have that MP is working or will work given this environment not about whether we can say monetary policy works in general.
I have some pretty wild graphs on everything from sovereign debt to money supply with respect to the Japanese attempts to use monetary policy (and fiscal policy) to stimulate out of the slump. In short, the time series go exponential while GDP stayed at near flat-line. I suppose we could go counter factual and speculate what GDP growth in Japan would have looked like in absence of these measures.
But to my mind the possibility that we face something like a low probability of success on the stimulus front regardless of whether that takes a monetary or fiscal form suggests, to my mind at least, policy ought to start with triage: getting money to low income and unemployed individuals and families and then work up or across from there.
So sure lower interest rates but there is no need to wait and see if that works before making transfers to the poor and unemployed.
That would be “pushing on a string”, as JMK said. But that’s conventional monetary policy in normal times.
Nick has made some important points on WCI that beyond interest rate reductions, the Bank of Canada can intervene by printing money and injecting it through the purchase of range of asset types. And purchases of Canadian government debt could be the way, if we got really desperate, that we would finance the fiscal stimulus.
Except that at the point at which the state is propping up asset prices through their purchase with government debt and printed money one has to wonder about the sentiment of investors.
It gets like chemotherapy on a close to acute patient: To cure him you must kill him as though the sword of Damacles hung above his head.
Was not the form I though revolution would take but then again better it happen by the “apolitical” central bank than in, what is now, a locked parliament.
Travis:
I think that RREMP is about as fair a caption as you can get in so few words!
To try to answer your question as best I can (with the RREMP always looking over my shoulder): I have mostly theoretical reasons that pure MP will not work, pure FP may or may not work, but money-financed FP (which is really MP+FP) will work at the zero lower bound (liquidity trap). I am bad at history (like it, and respect its importance, but my brain just can’t handle all the detail), so have to rely on others’ interpretation of Japan and the Depression. My understanding is that the evidence is not clear-cut, but at least does not contradict my theoretical priors.
But what happens when we get “close” to a liquidity trap? And how close is “close”? My hunch is that MP is currently less effective than normal (in Canada, I would give a different answer in US), but still effective. The risk premium is higher than normal, so risky assets are less close substitutes for short safe assets than they normally are. This should mean that the Bank has to cut the overnight rate by more than normal to get the same effect on aggregate demand. This is partly why I was the outlier in today’s CD Howe MPC decision, recommending the BoC cut from 2.25% to 1% immediately, which is a massive cut by normal standards. Empirically (be quiet, RREMP), I think I see the effect of the ‘surprise’ component of rate cuts still having an effect across the interest rate curve in Canada.
But yes, what is the counterfactual, that is the big question. Looking at ‘event studies’ of the surprise component of rate cuts is the only empirical evidence we have right now.
Some might say that that is a very weak set of evidence on which to base the future of our livelihoods, and they would be right. Except that the evidence for alternative policies is no better (there’s also an RREFP variant for fiscal policy), and theoretically, pure FP could conceivably go in either direction, i.e. it’s theoretically possible for a fiscal stimulus to make things worse (and Mankiw’s blog cites empirical evidence which suggests it might sometimes do just that).
Now you can agree or disagree with whether the BoC has and will cut interest rates far enough, but as long as they are above zero, you have to recognise that the BoC is in charge of AD (whether you like it or not), and that if the BoC believes that interest rates are right, then the only effect of a pure FP would be for the BoC to raise interest rates and/or the exchange rate to exactly offset any effect of pure fiscal stimulus on AD. That’s the nature of the game between fiscal and monetary authorities. So even if pure FP does work, ceteris paribus, other things, namely monetary policy, will not be equal.
Let’s talk about the elephant for a little: ideology. I’m right, most of you guys are left (I think). I can live with that, as I’m sure most of you guys can too (some of my best friends are…). There are two ideological issues:
1. Most on the left want some increased government spending or transfers for ‘microeconomic’ reasons; you would want them even if we weren’t in macro danger. Most on the right, the opposite. OK, we disagree. But I think we should have that debate on micro grounds, not muddy the waters with macro reasoning. (I wish I could express this more clearly.) If we do hit the zero lower bound, we will have a rare coincidence (a marriage of convenience) where we both want more or less the same thing. But it is just a coincidence.
2. This is the biggy. While the left has credibility on micro FP, about equal to the right, the left has zero credibility on MP. Here’s why. I cannot ever remember anyone on the left ever arguing for higher interest rates over the last 15 years of inflation targeting (or before for that matter). But on average, over the last 15 years, the rate of inflation has been almost exactly at the 2% target. So on average the BoC was exactly right (which doesn’t mean it was right the whole time of course). And anyone who argued that the BoC should lower rates, but never said it should raise rates, is therefore wrong. (I really ought to write a post on WCI on this topic.) It would be good for the left, and I admit, good for everybody, if the left had credibility on MP too. But that can only happen if the left sometimes recommends the BoC raise rates, and sometimes complains that BoC is setting rates too low. Sorry for being a bit confrontational on your blog on this (narrow) question, but it’s important for all of us. E.g., the CAW (though I don’t like to admit it, and have some real issues with them on other grounds) might actually know something about the general state of the economy sometimes (and Jim Sanford (really sorry if the spelling is wrong)) is certainly more than economically literate. But zero credibility on MP, given past experience.
Thanks Marc!
No I must shut up, and go and cook dinner. Thanks for reading my ramblings.
Hi Nick,
Thanks for contributing. It is not so much as a left-right thing for me on the blog as it is about having a good discussion and this post has delivered on that front.
Up until recently my sentiment has been that the Bank of Canada has been too hawkish in its inflation stance. We had a lengthy discussion about this one with Stephen Gordon a while back so I don’t want to get it rolling again. Perhaps if we engage in a massively expansive fiscal policy and we err on the side of doing too much I will have an occasion to call for higher interest rates!
Ok I will bite, but only if we all play by the same rules (that is I get access to factuals and counter factuals)
“But on average, over the last 15 years, the rate of inflation has been almost exactly at the 2% target. So on average the BoC was exactly right (which doesn’t mean it was right the whole time of course).”
That would be the RREMP in action. Sure but it may be that the “natural rate of inflation” (oh shit I just said that) was the stronger pole of attraction so that interest rates could have been lower and still produced the same or nearly the same inflation rate with a lower unemployment rate. I think the empirical record on the NAIRU demonstrates that there is hardly a one to one correspondence btw the rate of inflation and the unemployment rate. Ok so that is my counter-factual based on a empirical relationship between unemployment inflation and interest. Which by extension I would then argue that the statement that BOC got it exactly right is a little strong. I would take in the ball-park.
“And anyone who argued that the BoC should lower rates, but never said it should raise rates, is therefore wrong.”
Well there were some of us who did bulk at using cheap money in place of fiscal policy to stimulate particularly with respect to the US but also with respect to Canada. And quite frankly it has been the right, (conservatives whatever you want to call them) who have been in control of interest rates and the last 7 years are hardly edifying for them in terms of monetary policy and its broader implications for global and national economic stability.
I would help I think if we thought rather in terms of inflationary environments (conjunctures, cycles) rather than in terms of a fixed relationship between CB interest rates and inflation. Arguably that could put me at Greenspan circa early 2000s save for the fact I would have not argued that low inflation was a green-light for loose money. Let us call that the MP conceit: As long as inflation is good to hell with everything else.
So let me turn it around and argue that it would be good for the right, and I admit good for everybody, if the right admitted it has lost credibility on monetary policy.
And on my part I do not find any of this confrontational. Was a time when it was called the hard business of debate and learning. So I thank you for the provocation and education.
Should have read lost *some* credibility on monetary policy.