National decides to build a giant statue of John Key at a cost of $4 billion, to be paid for out of new borrowing. The project requires 25,000 workers for one year.

What would be the effect on inflation? There would almost certainly be some inflationary pressure with a project of that size. However, the effects on inflation of stimulus spending in other countries during this crisis has been small.

What would be the effect on interest rates? The Reserve Bank would likely raise the cash rate a notch or two. A hawkish governor would likely seize the moment to initiate a series of rate rises.

What would be the effect on the bond market? If the market believes the statue is a one-off, the effect should be small: of the order of the Reserve Bank's change in the OCR. If it suspects that every Prime Minister is going to want a $4 billion statue now, the effect would be large.

What would be the short-term effect on growth? The newly employed would have money to spend, which would have a multiplier effect. Working against this would be any increase in interest rates. GDP for the year would probably be boosted by less than $4 billion, but probably not much less than $4 billion unless the Reserve Bank was hawkish.

What would be the short-term effect on employment? Employment would be boosted by less than $25,000, but probably not much less than 25,000.

What would be the medium-term effect on the economy? Most of the boosts to GDP and employment would persist.

What would be the medium-term effect on public finances? Tax receipts would be boosted. Would they be boosted enough to compensate for increased debt servicing costs? It'd be a close-run thing.

Is building a $4 billion statue of John Key a good idea? It's not quite worth the uncertainty. Spending money on something productive is a different matter though.

Permalink | Leave a comment  »

[update: edited to add vertical lines for elections]

R code after the jump. Things to fix: Currently the lines don't go through the actual election results. However, just upweighting the election results in the Loess gives weird results. In fact I kind of think Loess isn't quite right here. Use another smoother?

###Code starts here

rm(list=ls())

#==========================================

#####Parameters

selected.parties <- c("Green","Labour","National", "NZ First")   #use precise names from Table headers

start = ISOdate(2005,9,1)

finish = ISOdate(2012,6,1)

#==========================================

#Misc preparation

selected.parties <- gsub(" ","_",selected.parties)  #Handle the space in some names

#==========================================

#Get 2011-2014 data

#Load the complete HTML file into memory

html <- readLines(url("http://en.wikipedia.org/wiki/Opinion_polling_for_the_next_New_Zealand_general_election",encoding="UTF-8"))

closeAllConnections()

#Extract the opinion poll data table

tbl.no <- 2

tbl <- html[(grep("<table.*",html)[tbl.no]):(grep("</table.*",html)[tbl.no])]

#Now split it into the rows, based on the <tr> tag

tbl.rows <- list()

open.tr <- grep("<tr",tbl)

close.tr <- grep("</tr",tbl)

for(i in 1:length(open.tr)) tbl.rows[[i]] <- tbl[open.tr[i]:close.tr[i]]

#Extract table headers

hdrs <- grep("<th",tbl,value=TRUE)

hdrs <- hdrs[1:11] # Manual fudge

party.names <- gsub("<.*?>","",hdrs)[-c(1:2)]

party.names <- gsub(" ","_",party.names)  #Replace space with a _

party.names <- gsub("M.{1}ori","Maori",party.names)  #Apologies, but the hard "a" is too hard to handle otherwise

party.cols   <- gsub("^.*bgcolor=\"(.*?)\".*$","\\1",hdrs)[-c(1:2)]

names(party.cols) <- party.names

#Extract data rows

tbl.rows <- tbl.rows[sapply(tbl.rows,function(x) length(grep("<td",x)))>1]

#Now extract the data

survey.dat <- lapply(tbl.rows,function(x) {

  #Start by only considering where we have <td> tags

  td.tags <- x[grep("<td",x)]

  #Polling data appears in columns 3-11 #fudge?

  dat     <- td.tags[3:11]

  #Now strip the data and covert to numeric format

  dat     <- gsub("<td>|</td>","",dat)

  dat     <- gsub("%","",dat)

  dat     <- gsub("-","0",dat)

  dat     <- gsub("<","",dat)

  dat     <- as.numeric(dat)

  names(dat) <- party.names

  #Getting the date strings is a little harder. Start by tidying up the dates

  date.str <- td.tags[2]                        #Dates are in the second column

  date.str <- gsub("<sup.*</sup>","",date.str)   #Throw out anything between superscript tags, as its an reference to the source

  date.str <- gsub("<td>|</td>","",date.str)  #Throw out any tags

  #Get numeric parts of string

  digits.str <- gsub("[^0123456789]"," ",date.str)

  digits.str <- gsub("^ +","",digits.str)    #Drop leading whitespace

  digits     <- strsplit(digits.str," +")[[1]]

  yrs        <- grep("[0-9]{4}",digits,value=TRUE)

  days       <- digits[!digits%in%yrs]

  #Get months

  month.str <- gsub("[^A-Z,a-z]"," ",date.str)

  month.str <- gsub("^ +","",month.str)        #Drop leading whitespace

  mnths     <- strsplit(month.str," +",month.str)[[1]]

  #Now paste together to make standardised date strings

  days  <- rep(days,length.out=2)

  mnths <- rep(mnths,length.out=2)

  yrs <- rep(yrs,length.out=2)

  dates.std <- paste(days,mnths,yrs)

#  cat(sprintf("%s\t -> \t %s, %s\n",date.str,dates.std[1],dates.std[2]))

  #And finally the survey time

  survey.time <- mean(as.POSIXct(strptime(dates.std,format="%d %B %Y")))

  #Get the name of the survey company too

  survey.comp <- td.tags[1]

  survey.comp <- gsub("<sup.*</sup>","",survey.comp)

  survey.comp <- gsub("<td>|</td>","",survey.comp)

  survey.comp <- gsub("<U+2013>","-",survey.comp,fixed=TRUE)

  survey.comp <- gsub("(?U)<.*>","",survey.comp,perl=TRUE)

  #And now return results

  return(data.frame(Company=survey.comp,Date=survey.time,date.str,t(dat)))

})

#Combine results

surveys14 <- do.call(rbind,survey.dat)

#==========================================

#Get the 2008-2011 data

#Load the complete HTML file into memory

html <- readLines(url("http://en.wikipedia.org/wiki/Opinion_polling_for_the_New_Zealand_general_election,_2011",encoding="UTF-8"))

closeAllConnections()

#Extract the opinion poll data table

tbl.no <- 2

tbl <- html[(grep("<table.*",html)[tbl.no]):(grep("</table.*",html)[tbl.no])]

#Now split it into the rows, based on the <tr> tag

tbl.rows <- list()

open.tr <- grep("<tr",tbl)

close.tr <- grep("</tr",tbl)

for(i in 1:length(open.tr)) tbl.rows[[i]] <- tbl[open.tr[i]:close.tr[i]]

#Extract table headers

hdrs <- grep("<th",tbl,value=TRUE)

hdrs <- hdrs[1:(length(hdrs)/2)]

party.names <- gsub("<.*?>","",hdrs)[-c(1:2)]

party.names <- gsub(" ","_",party.names)  #Replace space with a _

party.names <- gsub("M.{1}ori","Maori",party.names)  #Apologies, but the hard "a" is too hard to handle otherwise

party.cols   <- gsub("^.*bgcolor=\"(.*?)\".*$","\\1",hdrs)[-c(1:2)]

names(party.cols) <- party.names

#Extract data rows

tbl.rows <- tbl.rows[sapply(tbl.rows,function(x) length(grep("<td",x)))>1]

#Now extract the data

survey.dat <- lapply(tbl.rows,function(x) {

  #Start by only considering where we have <td> tags

  td.tags <- x[grep("<td",x)]

  #Polling data appears in columns 3-11

  dat     <- td.tags[3:12]

  #Now strip the data and covert to numeric format

  dat     <- gsub("<td>|</td>","",dat)

  dat     <- gsub("%","",dat)

  dat     <- gsub("-","0",dat)

  dat     <- gsub("<","",dat)

  dat     <- as.numeric(dat)

  names(dat) <- party.names

  #Getting the date strings is a little harder. Start by tidying up the dates

  date.str <- td.tags[2]                        #Dates are in the second column

  date.str <- gsub("<sup.*</sup>","",date.str)   #Throw out anything between superscript tags, as its an reference to the source

  date.str <- gsub("<td>|</td>","",date.str)  #Throw out any tags

  #Get numeric parts of string

  digits.str <- gsub("[^0123456789]"," ",date.str)

  digits.str <- gsub("^ +","",digits.str)    #Drop leading whitespace

  digits     <- strsplit(digits.str," +")[[1]]

  yrs        <- grep("[0-9]{4}",digits,value=TRUE)

  days       <- digits[!digits%in%yrs]

  #Get months

  month.str <- gsub("[^A-Z,a-z]"," ",date.str)

  month.str <- gsub("^ +","",month.str)        #Drop leading whitespace

  mnths     <- strsplit(month.str," +",month.str)[[1]]

  #Now paste together to make standardised date strings

  days  <- rep(days,length.out=2)

  mnths <- rep(mnths,length.out=2)

  yrs <- rep(yrs,length.out=2)

  dates.std <- paste(days,mnths,yrs)

#  cat(sprintf("%s\t -> \t %s, %s\n",date.str,dates.std[1],dates.std[2]))

  #And finally the survey time

  survey.time <- mean(as.POSIXct(strptime(dates.std,format="%d %B %Y")))

  #Get the name of the survey company too

  survey.comp <- td.tags[1]

  survey.comp <- gsub("<sup.*</sup>","",survey.comp)

  survey.comp <- gsub("<td>|</td>","",survey.comp)

  survey.comp <- gsub("<U+2013>","-",survey.comp,fixed=TRUE)

  survey.comp <- gsub("(?U)<.*>","",survey.comp,perl=TRUE)

  #And now return results

  return(data.frame(Company=survey.comp,Date=survey.time,date.str,t(dat)))

})

#Combine results

surveys11 <- do.call(rbind,survey.dat)

#==========================================

#Get the 2005-2008 data

#Load the complete HTML file into memory

html <- readLines(url("http://en.wikipedia.org/wiki/Opinion_polling_for_the_New_Zealand_general_election,_2008"),encoding="UTF-8")

closeAllConnections()

#Extract the opinion poll data table

tbl.no <- 2

tbl <- html[(grep("<table.*",html)[tbl.no]):(grep("</table.*",html)[tbl.no])]

#Now split it into the rows, based on the <tr> tag

tbl.rows <- list()

open.tr <- grep("<tr",tbl)

close.tr <- grep("</tr",tbl)

for(i in 1:length(open.tr)) tbl.rows[[i]] <- tbl[open.tr[i]:close.tr[i]]

#Extract table headers

hdrs <- grep("<th",tbl,value=TRUE)

hdrs <- hdrs[1:10] #fudge

party.names <- gsub("<.*?>","",hdrs)[-c(1:2)]

party.names <- gsub(" ","_",party.names)  #Replace space with a _

party.names <- gsub("M.{1}ori","Maori",party.names)  #Apologies, but the hard "a" is too hard to handle otherwise

#party.cols   <- gsub("^.*bgcolor=\"(.*?)\".*$","\\1",hdrs)[-c(1:2)]

#names(party.cols) <- party.names

#Extract data rows

tbl.rows <- tbl.rows[sapply(tbl.rows,function(x) length(grep("<td",x)))>1]

#Now extract the data

survey.dat <- lapply(tbl.rows,function(x) {

  #Start by only considering where we have <td> tags ### or <th>

  td.tags <- x[grep("<td|<th",x)]

  #Polling data appears in columns 3-10 #fudge

  dat     <- td.tags[3:10]

  #Now strip the data and covert to numeric format

  dat     <- gsub("<td>|</td>","",dat)

  dat     <- gsub("%","",dat)

  dat     <- gsub("-","0",dat)

  dat     <- gsub("<","",dat)

  dat     <- as.numeric(dat)

  names(dat) <- party.names

  #Getting the date strings is a little harder. Start by tidying up the dates

  date.str <- td.tags[2]                        #Dates are in the second column

  date.str <- gsub("<sup.*</sup>","",date.str)   #Throw out anything between superscript tags, as its an reference to the source

  date.str <- gsub("<td>|</td>","",date.str)  #Throw out any tags

  ### Get rid of "Released "

  date.str <- gsub("Released ","",date.str)

  ### Throw out anything italicized

  date.str <- gsub("<i.*</i>","",date.str)

  date.str <- gsub("&#160;"," ",date.str)

  #Get numeric parts of string

  digits.str <- gsub("[^0123456789]"," ",date.str)

  digits.str <- gsub("^ +","",digits.str)    #Drop leading whitespace

  digits     <- strsplit(digits.str," +")[[1]]

  yrs        <- grep("[0-9]{4}",digits,value=TRUE)

  days       <- digits[!digits%in%yrs]

  #Get months

  month.str <- gsub("[^A-Z,a-z]"," ",date.str)

  month.str <- gsub("^ +","",month.str)        #Drop leading whitespace

  mnths     <- strsplit(month.str," +",month.str)[[1]]

  #Now paste together to make standardised date strings

  days  <- rep(days,length.out=2)

  mnths <- rep(mnths,length.out=2)

  yrs <- rep(yrs,length.out=2)

  dates.std <- paste(days,mnths,yrs)

#  cat(sprintf("%s\t -> \t %s, %s\n",date.str,dates.std[1],dates.std[2]))

  #And finally the survey time

  survey.time <- mean(as.POSIXct(strptime(dates.std,format="%d %B %Y")))

  #Get the name of the survey company too

  survey.comp <- td.tags[1]

  survey.comp <- gsub("<sup.*</sup>","",survey.comp)

  survey.comp <- gsub("<td>|</td>","",survey.comp)

  survey.comp <- gsub("<U+2013>","-",survey.comp,fixed=TRUE)

  survey.comp <- gsub("(?U)<.*>","",survey.comp,perl=TRUE)

  #And now return results

  return(data.frame(Company=survey.comp,Date=survey.time,date.str,t(dat)))

})

#Combine results

surveys08 <- do.call(rbind,survey.dat)

### Let's put it all together

# Make 2008 column names consistent

names(surveys08) = c("Company","Date","date.str","Labour","National","NZ_First","Maori","Green","ACT","United_Future","Progressive")

# Avoid doubling up on the elections

surveys08 = surveys08[-nrow(surveys08),]

surveys11 = surveys11[-nrow(surveys11),]

# There must be a better way to do the merging than this:

surveys = merge(surveys08,merge(surveys11,surveys14,all=T),all=T)

surveys=surveys[order(surveys$Date),]

### Draw the graph

#Restrict plot(manually) to selected parties

selected.parties <- sort(selected.parties)

selected.cols <- party.cols[selected.parties]

polls   <- surveys[,c("Company","Date",selected.parties)]

polls <- subset(polls,!is.na(surveys$Date))

polls <- polls[order(polls$Date),]

polls$date.num  <- as.double(polls$Date)

xlims <- c(start,finish)

ticks <- ISOdate(2005:2013,1,1)

par(mar=c(5,4,1,1))

matplot(polls$date.num,polls[,selected.parties],pch=NA,xlim=xlims,ylab="Party support (%)",

xlab="",xaxt="n",yaxs="i",col=selected.cols)

abline(h=seq(0,95,by=5),col="lightgrey",lty=3)

abline(v=as.double(ticks),col="lightgrey",lty=3)

box()

axis(1,at=as.double(ticks),labels=format(ticks,format="1 %b\n%Y"),cex.axis=0.8)

axis(4,at=axTicks(4),labels=rep("",length(axTicks(4))))

#Now calculate the loess smoothers and add the confidence interval

smoothed <- list()

predict.x <- seq(min(polls$date.num),max(polls$date.num),length.out=100)

for(i in 1:length(selected.parties)) {

  smoother <- loess(polls[,selected.parties[i]] ~ polls[,"date.num"],span=0.5)

  smoothed[[i]] <- predict(smoother,newdata=predict.x,se=TRUE)

  polygon(c(predict.x,rev(predict.x)),

    c(smoothed[[i]]$fit+smoothed[[i]]$se.fit*1.96,rev(smoothed[[i]]$fit-smoothed[[i]]$se.fit*1.96)),

    col=rgb(0.5,0.5,0.5,0.5),border=NA)

}

names(smoothed) <- selected.parties

#Then add the data points

matpoints(polls$date.num,polls[,selected.parties],pch=20,col=selected.cols)

#And finally the smoothers themselves

for(i in 1:length(selected.parties)) {

  lines(predict.x,smoothed[[i]]$fit,lwd=2,col=selected.cols)

}

# Manually add election lines

abline(v = ISOdate(2011,11,26), lwd=3)

abline(v = ISOdate(2008,11,8), lwd=3)

abline(v = ISOdate(2005,9,17), lwd=3)

legend("bottom",legend=gsub("_"," ",selected.parties),col=selected.cols,pch=20,bg="white",lwd=2,horiz=TRUE,inset=-0.225,xpd=NA)

#Add best estimates

for(i in 1:length(smoothed)) {

  lbl <- sprintf("%2.0f±%1.0f %%",round(rev(smoothed[[i]]$fit)[1],0),round(1.96*rev(smoothed[[i]]$se.fit)[1],0))

  text(rev(polls$date.num)[1],rev(smoothed[[i]]$fit)[1],labels=lbl,pos=4)

}

cat("Complete.\n")

Permalink | Leave a comment  »

Experimental studies on teacher performance pay:

POINT Experiment, Nashville

New York City Bonus Program

Andhra Pradesh

There's also a quasi-experimental study from Israel I haven't looked for.

My best guess is that performance pay is effective in the developing world but not the developed world.

Permalink | Leave a comment  »

The kind of leftism I prefer supports as the first-order problem an increase in the size and scope of government. Essentially that means more tax, at the right time. The efficiency and progressivity of the tax system also matter, but they're second-order concerns. If National's first-term tax changes had been implemented some time other than the aftermath of a financial crisis, and if they had truly been revenue-neutral (cue laugh track), I would've only been slightly grumpy about their regressive structure. Income taxes do have *some* drag, so I'd prefer a tax on capital gains (or capital, or just land), but really, the point is to be prepared to hike taxes one way or another once we're out of stagnation.

Permalink | Leave a comment  »

The Reserve Bank began cutting the cash rate early, in July 2008, at a time when there was still inflationary pressure (core inflation in the threes). As Lehman and friends fell, Bollard was not afraid to deliver big cuts. By the end of April 2009, the OCR had fallen from 8.25% to 2.5%. In contrast, Australia's cash rate only dropped from 7.25% to 3%, though the necessity for low rates there was lessened by aggressive fiscal policy. Usefully, Bollard didn't rule out the possibility of further cuts, which helped keep the banks in line, while his belief that inflation would be brought back into his target zone by the recession was vindicated.

Bollard put the rate up a quarter point in June 2010. The move was widely expected, with the Bank expecting 3.5% growth and a commodity-driven inflation wave on the horizon*, but some of us (e.g. Kel Sanderson) thought it was a grave mistake. The hike itself was bad enough, but the lack of clarity as to how high rates could go perhaps also damaged the fragile recovery -- banks seemed to think the OCR would rise to 5-6%. Instead, there was one further hike before the Christchurch earthquakes happened. I believe the rises would have had to be stopped or reversed in any case, but the earthquakes made it inarguable. The OCR returned to 2.5%, and should be there for a while.

So Bollard had a questionable call in 2006, a great call in 2007, and a bad call in 2010, and was good the rest of the time. A key question to address before appointing his successor is what led to the bad call. One contributing factor was the inaccurate growth forecasts in the aftermath of the crisis. While the Reserve Bank was far from the worst offender in this respect, their errors were nevertheless costly. Another factor was the pre-emptive reaction to the forecasted commodity bubble of late 2010/early 2011. In this case, the forecast was correct. Yet the inflation was transitory: the CPI and all measures of core inflation are now well within the target zone. Part of the problem was that some of the core inflation measures didn't seem to know the commodity bubble was transitory, while another part was the target was too low given the fragility of the economy. A rethink of the policy target agreement before Bollard's successor is chosen would be wise.

*edit: and a GST increase, duh

Permalink | Leave a comment  »

Bollard has confirmed he won't stand for another term when the current one ends in September. Almost everyone in a position of great power ends up with a mixed record, and Bollard is no exception.

Bollard took over from Brash just as rapid growth in house prices was starting. For the first couple of years, there was good reason to believe this was just compensating for several preceding years of weak house price growth. By 2004, however, there was reason to ask whether the housing market was becoming a bubble. Bollard started hiking the cash rate, though he denied it was because of housing.

While house price growth slowed from its absurd 2003 peak, it remained in double-digits. On the other hand, exporters were beginning to suffer from tight money. Bollard left the cash rate at 7.25% for all of 2006. It's easy to say in retrospect this was the wrong call: that he should have kept hiking rates and intervened directly in the currency market. But at the time the discussion was more about when rates would fall than whether they would rise again. The fact is that his stated job was to control overall inflation, and while he saw there were imbalances that would require adjustment, it wasn't his brief.

In 2007, however, the coming adjustment was staring us in the face. Bollard resumed rate hikes and sold NZD, finally taking the heat out housing. He did this in the face of substantial opposition from business leaders and unions, as well as National. Cullen was worried, even proposing a mortgage levy (which Clark shot down), but fiscal policy became increasingly stimulative. This was arguably Bollard's shining moment, and almost certainly softened the impact of the financial crisis. It would have been even better had he brought the hammer down on the banks, but this would have required remarkable foresight. As 2008 began and the prospect of an international recession became strong, Bollard continued to be hawkish.

(to be continued)

Permalink | Leave a comment  »

In some ways, the recession NZ suffered wasn't that bad. The 1990-1991 crunch, for instance, saw a larger bounce in unemployment. Yet in that case, recovery was clearly under way by the 1993 election, saving Bolger's bacon. This time around there are only the barest hints of recovery. Why?

Explanation 1: The problems in the early '90s were specific to NZ, not global. There were always overseas markets to sell to. This time around all the developed economies have been hit, many worse than us.

Explanation 2: Financial crises are worse than other kinds. They leave a lot of people with substantial debt problems, making it hard to restart the economy until deleveraging has happened or evasive action is taken.

Both of these explanations are true, but is one truer than the other? Financial crises tend to be international, so it's hard to separate out their effects. The policy implications of the two views are somewhat different, though not vastly so (1 requires more internation co-ordination than 2).

Permalink | Leave a comment  »

The Economist has an article comparing deleveraging over a bunch of countries since the crisis. According to a McKinsey report, while the US and South Korea have reduced total debt as a percentage of GDP, in many countries, notably Japan, France, and Spain, total debt has soared.

How's New Zealand doing? Let's use Reserve Bank figures, and stick to foreign debt. At the end of 2005, total foreign debt was 107%, composed of 96% corporate and 11% government debt. This seems like the level we should be shooting for. This soared to 138% in March 2009, all of the increase coming in corporate debt. We got this down to 127% in June 2011, but the following bad-for-everyone quarter saw this rise to 134% (111% corporate, 22% government). One message is that one bad quarter (that occurred for reasons outside NZ's control) can undo two years of good work -- yet another reason to fear Eurodefault. Still, the overall process of increasing government debt while the private sector deleverages seems to be sound. Absent any further shocks (ha), the private sector should be back down to pre-bubble debt levels in a couple of years. The question, though, is whether confidence has been so shaken that even then we won't have returned to our long-run path.

Permalink | Leave a comment  »

Mitt Romney is close to wrapping up the Republican presidential nomination. If he wins in South Carolina, where he has an eight-point lead over Newt Gingrich, he's essentially proved he can win anywhere and the party elite will move to bring everyone behind the presumptive candidate. The opposition to Romney from the right has been disorganised, unable to pick a pretender and stick with him. Failing to squish Ron Paul is one thing, but still having Gingrich and Rick Santorum squabble over true-conservative votes shows poor judgement. Gingrich has been reduced to going negative on Romney, yet evangelical endorsements are still going to Google-bombed Santorum. Democrats must be hoping for an unlikely upset in South Carolina or Florida to keep the race alive.

More likely, Romney will wrap it up quickly and it'll be time to handicap his matchup with Obama more precisely. Neither is especially popular compared to recent nominees, with favourable and unfavourable ratings roughly matched. In a referendum-style election, this would look worse for the incumbent. The betting markets, however, give Obama a slight edge, which might seem counter-intuitive -- especially with a stagnant economy. It's not just the current economy that explains election results, however, it's also the improvement in the economy over the term, which helps out governments around the world that camne to power just after the financial crisis.

The presidential race currently looks like it will be tighter than the battles for Senate and House control. The Republicans have a huge structural advantage in the year's Senate races, and it will be surprising if they don't pick up a majority. In the House, the Democrats have a fighting chance, but they have a huge deficit to make up in an incumbent-friendly system. Their best chance will be if Obama has a clear advantage over Romney on election day, and brings in a bunch of Democrats on his coattails. While the US left's love affair with Obama is on the rocks, he remains their best avenue for electoral success.

Permalink | Leave a comment  »

The pressing task for Labour is to construct a compelling critique of National. They pretty much failed to do this last term. The critique then was about stuff that National had yet to do, like asset sales, rather than past performance — aside from complaining about the deficit, which was never a winning line. Seems to me there’s political hay to be made in pointing out that growth under Key has been so much lower as to be nonexistent, regardless of why. Estimates vary, but the Bolger/Shipley years saw something like 3.2% average annual growth, while the Clark government presided over an average of 2.9% (they were beating their predecessors until the disastrous 2008). Under Key it's been 0.7%; per capita, we're still below our pre-crisis level.

National would like to be associated with growth, but you can make a strong case they don't care about pursuing it ahead of other economic goals. Where's the evidence to the contrary? (If you say “tax cuts”, you’re welcome to try to convince me that their primary aim wasn’t to redivide the pie.) The Nats seem perfectly satisfied stagnating along, and why wouldn’t they be when the public hasn’t called them on it? In fact, last year they pretty much stopped talking about catching up to Australia (smartly, since Australia have moved even further ahead post-crisis), instead telling us about how we were going to return to surplus soon, it must be true, Treasury said so.

In the short term any growth should be achievable with minimal negative effects because there’s so much slack in the economy. Growth would put people into work, rather than out of it (except maybe Alan Bollard). If all Labour's “growing the pie” bullshit leads to a coherent critique of National’s performance, then good for them (and good for us, if their nagging spurs National into actually doing something to end the stagnation). The use of cliches suggests they haven’t figured out how to grow the pie yet — aside from the all-purpose answer of SCIENCE! — but as they’re not going to be in power for at least three years, and because the last election showed voters are indifferent to Opposition policy details, this isn’t as urgent.

Permalink | Leave a comment  »

In 2009, in the wake of the financial crisis, the Reserve Bank dropped the OCR to 2.5%. It rose to 3% for a while before it returned to 2.5% after the Christchurch earthquakes; it looks like there will be no more hikes for some time yet. Meanwhile, unemployment has bounced between 6% and 7% since mid-2009, with no clear downward trend. Textbook macro says you can stimulate the economy and lower unemployment by lowering interest rates. Yet there has been little argument for lowering the cash rate further. Why?

The main reason is that has been some inflation, starting with the commodity bubble of late 2010. Annual CPI inflation is currently 4.6%. What core inflation is depends on your preferred definition of core inflation, but the majority of the Reserve Bank's measures put it above 3%, the maximum allowed by the policy target agreement.

What risks would be entailed by loosening of the inflation target? The chance of accelerating inflation is essentially zero. You could grumble about loss of central bank independence, but it's hard to see any loss of credibility caused by bounded loosening in a global economic environment as extreme as this.

Conspirators may note that many of National's donors may be quite happy with a labour market in which high cyclical unemployment turning into high structural employment. I think the explanation is simpler. The Key Government is a small-c conservative government. Its huge poll leads have meant that the safest course of action, in terms of getting re-elected in 2011, was to do as little as possible. Meanwhile, there were no votes for Opposition parties in monetary policy. David Parker finally announced a half-assed monetary policy a couple of weeks before the election, to no response.

The problem for Key is that if he wants a third term, he needs a recovery. One weak quarter a year, like the 0.1% growth in 2Q 2011, is enough to wreck this. Key may have to consider action soon.

Permalink | Leave a comment  »

Suppose that your primary macroeconomic worry is the public debt. While NZ has a low public debt and the bond rate is at a record low, the possibility of another global fuck-up poses a risk, and you might want to flatten NZ's debt trajectory to reduce that risk. Fiscally, National aren't going to enact further stimulus, let alone socialisation.Three kinds of options are under discussion in the consensusphere:

1. Reduce the deficit through tax hikes or lower spending
2. Reduce current debt by selling assets
3. Do nothing

In a weak economy, any action to reduce the deficit will be painful. Even cut in spending with a small proportionate labour cost, like motorways, would lead to substantial job losses. Firstly, spending on materials creates jobs in those sectors as well, albeit less directly. Secondly, a big chunk of the job creation or loss is through multiplier effects. Even a more left-correct option like a higher top tax rate would dent growth -- and thus might be counterproductive. You'd probably need a multiplier well below 0.5 for the tax hike to pay off even in strict terms of debt sustainability.

If reducing debt by selling assets seems like kicking the can down the road, it is. Since the bond rate is lower that the rate of profit, to improve the fiscal situation you need buyers that believe they can run the show more efficiently than the government. However, if you're not selling a majority stake, there doesn't seem to be any reason to believe efficiency gains are possible.

That leaves doing nothing. Like the other options, this relies on an eventual return to trend growth leading to a return to surplus. Thus the problem reduces, once again, to how to get the economy to or above trend growth sooner rather that later, which is the problem I usually write about. Even if you're primarily worried worried about debt, you should be primarily worried about growth.

Permalink | Leave a comment  »

The original sin of the euro crisis has much more to do with balance of payments problems than fiscal problems. This is a worry for NZ, since we have a huge structural balance of payments deficit -- although we've had a surplus in 2011 due to insurance inflows from the Christchurch earthquakes. While we have a major backstop in that we have our own currency, it's worth looking at the eurozone's monetary response in detail.

Pre-crisis, the current account deficits of the PIIGS were financed by voluntary lending, primarily from German banks. The crisis meant not only that this lending dried up, but capital flight as well. Instead, deficits have been indirectly financed through national central banks, effectively making everyone the Bundesbank's bitch. But the Bundesbank takes the risk that some countries may default. Furthermore, it has little direct control over German monetary conditions. As contagion spreads, it becomes increasingly important the ECB doesn't fuck up the short-term challenge of avoiding default or the medium-term challenge of adjusting balance of payments.

NZ is in excellent fiscal shape, but if the balance of payments is the driver then things aren't rosy. NZ's structural balance of payments deficit exists because foreigners have more capital in NZ than NZers have capital overseas. A big part of this consists of net liabilities to the Australian parents of NZ banks. The good news is that about 55% of our financial liabilities are denominated in NZ dollars. If things go way south, the RBNZ prints a bunch of money, the exchange rate drops, and a large proportion of the rebalancing is taken care of. Still, in a worst-case scenario where the Aussie parent banks are threatened, this probably won't be enough. Paying for bailouts would be painful, though probably not nearly bad enough to lead to default unless the policy response was totally bungled.

The RBNZ has plenty of ammunition to offset a second crisis. Still, it's worrying that the current account adjustment of 2008-09 has changed direction. One thing we should continue to have is higher inflation than Australia: another motivation for a looser policy target agreement.

Permalink | Leave a comment  »

Permalink | Leave a comment  »

Permalink | Leave a comment  »

Two recent papers -- Mian, Rao, and Sufi and Mian and Sufi -- have contained the kind of detailed empirical analysis of the Great Contraction that's been rare. The first shows that at the zip code level, high debt-to-income is associated with a fall in consumption, and that within counties, a high number of subprime borrowers is associated with a decline in auto sales. This is a decent test of the hypothesis that household debt has played a major role in the slump. The second finds that while tradeable sector job losses have been uniformly distributed, non-tradeable sector job losses have been greater in highly leveraged counties. This is consistent with the "fall in aggregate demand" story of the contraction, but is hard to square with supply-side theories like uncertainty or structural unemployment. It's not proof that the recession was demand-side (there are no proofs in empirical science), but it's a strong argument.

Their most challenging finding is that the shock to household balance sheets explains 65% of job losses in their data. It's the product of an economic model, which isn't gospel, but it implies that household debt is the heart of the problem. The question, then, is whether debt effects dominate any wealth effect over and above them. We know that the fall in house prices led to lower consumption. Is this mostly because banks were less willing to lend to homeowners? Or was it mostly because homeowners were less willing to spend? The answer matters because the two views imply different easiest ways out: direct debt relief vs fiscal stimulus.

Mian, Rao, and Sufi find the decline in consumption is much larger than previous studies on the wealth effect would suggest. This could also be explained by a "fear multiplier" making consumers much more cautious after 2008, however. Meanwhile, the incresae in the saving as a proportion of income has been comparatively small. That measure may be misleading, however, since real income shrunk. What we care about is the marginal saving rate.

An interesting experiment that will never happen would be to randomly select a bunch of people and give them $10K each, and see how much of that turns into consumption as a function of debt and wealth. Maybe use two groups: one that gets $10K cash and another that gets $10K of their mortgage paid off. The design isn't perfect -- we'd want some data from before the crisis to compare it to -- but it would be informative. Failing this, modelling observational data on consumption as a function of both debt and wealth would be useful but hard.

See also: Mike Konczal's interview with Amir Sufi.

Permalink | Leave a comment  »

No:

Fiscal space analysis purports to tell us how much a country can add to its (gross) public debt before the interest burden becomes unsustainable, if fundamentals remain constant. So the table says NZ could add 220% of its GDP to its public debt, which would give a total of 256%. These numbers, espeically the latter, shouldn't be taken literally. It's just a regression prediction, and it depends on inputs. The Moody's paper also gives a lower estimate of 178% fiscal space, or 214% total public debt, under a different interest rate assumption.

Furthermore, fiscal space can shrink rapidly when fundamentals change -- ask Iceland. But not only does NZ have a lot of fiscal space compared to other wealthy nations, it can survive a huge jump in bond rates. When fiscal space is calculated according to Moody's main set of assumptions, they recommend maintaining at least 125% fiscal space.

We're nowhere near the limit now. The calculations are interesting in that they give an indication of how much trouble we could be in if there's another crisis. The 125% rule suggests that NZ could deploy expansionary fiscal policy as long as gross debt peaked below 131%, assuming no decline in fundamentals. A moderate shock, like a Greek default, would leave us far short of the danger zone, implying we could spend our way out of recession. This doesn't mean that's the right policy, only that it wouldn't put us at significant risk of default. A huge shock, like complete disintegration of the eurozone, changes so much that the calculation has to be re-done. In any case, fiscal space analysis is not a substitute for analysis of the bond market. Fortunately, with the bond rate hovering around a record low 4%, the bond market is currently easy to analyse. So, again, the answer to the title question is no.

Permalink | Leave a comment  »

Permalink | Leave a comment  »

The eurozone lurches back toward a break-up.

The highest priority at the Brussels summit had to be to find the money to bail out Italy or Spain, should that become necessary. Without that money, there's no confidence -- a small ripple could cascade into a bond market collapse, and after the run starts, there won't necessarily be time to get the money together before contagion sets in.

The EFSF and its successor, the ESM, don't have the firepower to deal with a run on Italy or Spain. The ECB is currently content to sit on its hands (mustn't risk 4% inflation, now). I guess the hope is that if a run on a country starts, the ECB immediately comes out of its shell and snuffs it out before it spreads. Given this is exactly what didn't happen in Greece, one might have doubts.

How could the EU lower the risk of needing bailouts, without enlarging bailout funds? They're betting on austerity. (This might not achieve much in itself, but could give the ECB cover to act.) The trouble is that for any kind of fiscal union to have strength, it requires, at the least, all 17 eurozone states to sign on, and preferable the other 10 EU states as well. There doesn't seem to be much prospect of Britain chipping in, so the deal-making has largely been confined to the 17. But enforcing austerity in 17 out of 17 countries? Merkel might be able to foist it on her populace. But the peripheral countries won't be happy. Look at the resistance in Greece to austerity measures: you think you won't see the same thing in the rest of the PIGS? With inflation targets remaining low, you're asking countries to sign on to years of deflation to save the euro, and for a lot of the citizenry, that's not in their interest. That Merkel wants treaty changes only makes co-operation less feasible.

The best way out is still ECB money. In the absence of this, fiscal co-ordination could help, but it would require something that every single eurozone country could sign on to. The only way I can see this happening is if Germany accepts higher inflation. Failing that, any success would be indistinguishable from luck.

Permalink | Leave a comment  »

Though the news from the eurozone has been better this past week, it's clear that the risk of a euro break-up is high enough to necessitate thinking about contingency plans. So what should the Government and the Reserve Bank do if the unthinkable happens?

Of course this depends on the course of the break-up. If one or more of the PIGS leave, then while those holding money in the drop-outs would face devaluation and losses, those holding money in the reduced eurozone would win as the euro appreciated. It's not zero sum -- the losers would lose enough to cause some destabilisation. But in NZ, the ripples, though unpleasant, would probably be manageable. My guess is conventional monetary policy could handle it. Drop the cash rate to near zero, allow a little inflation but keep an eye out for bubbles. The government could help by explicitly condoning inflation up to, say, 4%.

The worse scenario is a total euro break-up, which could happen if Germany decide to take their ball and go home. The only certainty would be that lawyers would make a killing. My guess is that this would be worse than 2008. If so, NZ and Australia would have to co-ordinate policy. My short-term solution would be to nationalise a big chunk of the financial industry. Have some stimulus ideas bottled up, then release them as necessary as the crisis plays out.

National would never do this, which is one reason to be happy the Germans look willing to stay the course.

Permalink | Leave a comment  »