The transition has also preserved all comments. However, because the blog is an archive, I have now closed it to further comments.

At some point in the future, I’m hoping to resume blogging at andrewleigh.com.

The other candidates – Christina Ryan, Jim Jones, Michael Pilbrow, Mike Hettinger, Chris Bourke, Nick Martin and George Williams – are people for whom I have great respect. I was friends with most of them before the campaign began, and my admiration for them has only grown over past months.

I’m hoping to keep blogging, but have to think about the right way to evolve the blog (or whether to draw a line by starting a new blog in my capacity as an ALP candidate). So please forgive me while I try to sort that out.

In the meantime, I’m about to head to the US for what might be my last academic conference – the NBER economics of education meetings in Cambridge MA. I’ll be overseas from 26 April to 5 May. I haven’t seen Gweneth and my two little boys for nearly a month, and am missing them like crazy. Skype helps, but since the boys are aged 10 months and 3 years respectively, they don’t exactly want to talk to a computer screen for long.

For my wife Gweneth, this is going to be a particularly unusual experience. She passed her citizenship test last month, and becomes a dual Australian/US citizen on 4 June. So she’ll cast her first Australian ballot for her husband (if I’m lucky, that is..).

(a similar post was posted at Core Economics)

But it is not clear that boat people really had much effect on the election result. When the Tampa arrived, the Howard government had already been steadily improving its opinion poll position from the early 2001 nadir. Tampa and, later, “children overboard” melted the talkback lines, but so do lots of issues that don’t change votes.

It was September 11, two weeks later, that sent Howard’s voting-intentions figure skywards, but by polling day they had subsided, and the result of fifty-one to forty-nine, while respectable for a government in good economic times asking for a third term, was no landslide.

To answer the question, we need regular and precise estimates of who is likely to win the election. Unfortunately, polls are pretty irregular, and very imprecise (evidence here, anecdote here). But betting markets provide a stable, daily estimate. Here’s the time series for 2001, with the Tampa incident and September 11 attacks marked on the chart.

The chart is from this paper. To my mind, it doesn’t support Brent’s argument that the Howard government were sailing to victory before the Tampa incident. But it does accord with his view that the September 11 attacks were pretty important.

Increasing Time to Baccalaureate Degree in the United States by John Bound, Michael Lovenheim, Sarah Turner
Time to completion of the baccalaureate degree has increased markedly in the United States over the last three decades, even as the wage premium for college graduates has continued to rise. Using data from the National Longitudinal Survey of the High School Class of 1972 and the National Educational Longitudinal Study of 1988, we show that the increase in time to degree is localized among those who begin their postsecondary education at public colleges outside the most selective universities. In addition, we find evidence that the increases in time to degree were more marked amongst low income students. We consider several potential explanations for these trends. First, we find no evidence that changes in the college preparedness or the demographic composition of degree recipients can account for the observed increases. Instead, our results suggest that declines in collegiate resources in the less-selective public sector increased time to degree. Furthermore, we present evidence of increased hours of employment among students, which is consistent with students working more to meet rising college costs and likely increases time to degree by crowding out time spent on academic pursuits.

Education Policy and Crime by Lance Lochner
This paper discusses the relationship between education and crime from an economic perspective, developing a human capital-based model that sheds light on key ways in which early childhood programs and policies that encourage schooling may affect both juvenile and adult crime. The paper first discusses evidence on the effects of educational attainment, school quality, and school enrollment on crime. Next, the paper discusses evidence on the crime reduction effects of preschool programs like Perry Preschool and Head Start, school-age programs that emphasize social and emotional development, and job training programs for low-skill adolescents and young adults. Finally, the paper concludes with a broad discussion of education policy and its potential role as a crime-fighting strategy.

Financial Incentives and Student Achievement: Evidence from Randomized Trials by Roland G. Fryer
This paper describes a series of school-based randomized trials in over 250 urban schools designed to test the impact of financial incentives on student achievement. In stark contrast to simple economic models, our results suggest that student incentives increase achievement when the rewards are given for inputs to the educational production function, but incentives tied to output are not effective. Relative to popular education reforms of the past few decades, student incentives based on inputs produce similar gains in achievement at lower costs. Qualitative data suggest that incentives for inputs may be more effective because students do not know the educational production function, and thus have little clue how to turn their excitement about rewards into achievement. Several other models, including lack of self-control, complementary inputs in production, or the unpredictability of outputs, are also consistent with the experimental data.

I’m a tad distracted by other things at present, so apologies in advance for the fact that all three links are to the (gated) NBER website. Perhaps some kind soul will post ungated links in comments. The Fryer research is also in the latest issue of Time Magazine.

And if you still want more to read, Paul Krugman’s two new pieces on global warming and economic geography are masterful.

Their first competition is to build a model that predicts the results of Eurovision 2010. Given the large idiosyncratic component in this kind of contest, I’m not sure how well suited it is to prediction models (the model that best predicts the 2010 result is unlikely to be the model that best predicts the winners for 2010-2019). But it’s a clever way of getting the Kaggle idea out there quickly.

Statisticians of the world, unite. You have nothing to lose but your residuals.

(xposted @ Core Economics)

The Economics of Sleep, Australian Financial Review, 13 April 2010

Did you get enough sleep last night? According to time use surveys, the typical Australian gets about eight hours of sleep. But one-fifth of us kip for less than 6½ hours per night. Certain workers, such as bankers, truckers and new parents, often boast about how little sleep they can survive on. ‘Did you know I only sleep five hours a night?’, a colleague will say as she falls asleep in her espresso.

The standard economic perspective on sleep is that higher wages make sleep less attractive. The more you earn by putting in another hour on the job, the likelier you are to go to bed late and arise early. This helps explain why the rich sleep less than the poor, as well as why average hours of sleep have fallen over the past century. Moreover, the fact that wages rise over the lifecycle may also account for the fact that 25 year-olds get more sleep than 45 year-olds. Indeed, sleep is negatively associated with the business cycle, suggesting that more red eyes on Martin Place and Collins Street may point to a strong economy.

Should we care if people are cutting back on their shut-eye? There is clear medical evidence that too little sleep can be unhealthy. Physiologically, sleep allows the body to repair tissues and replenish hormones. Sleep deprivation has been linked to an array of health problems, including heart disease, diabetes, and weight gain. In one experiment, patients were given a vaccine, and then kept awake that night. A month later, their immune response was half that of patients who had a full night’s sleep after getting the vaccine.

However, if sleep deprivation were only about people opting to live longer days and fewer years, the rest of us should probably stand back and let them make that tradeoff. So the question is: does sleep deprivation also have adverse impacts on those around you?

To test the impact of sleep deprivation on workers’ productivity, an intriguing new strand of research takes volunteers into the laboratory and keeps them awake for 30-40 hours (the equivalent of pulling an all-nighter). They are then asked to complete tests alongside other subjects who are properly rested. In one experiment, sleep-deprived individuals were less willing to trust others. Another experiment indicated that sleep-deprived individuals did only about half as well on cognitive tests as their rested counterparts. Corroborating this, evidence from brain scans of sleepy and well-rested participants indicates that sleep-deprivation leads to less rational and more emotional behaviour.

In practical terms, employees who are untrusting, dim and over-emotional are likely to be bad for the firm’s bottom line. Much of business is about knowing who to trust, and carefully balancing decisions on their merits. Without enough sleep, you may miss key details, and make vital decisions based on gut instinct rather than careful reasoning.

Within the firm, sleep-deprived bosses who lose their temper with are likely to cause subordinates to quit or slack off. On the road, sleepy drivers have the reaction time of drunk drivers, and contribute to more crashes. Perhaps it is no surprise that major accidents such as Chernobyl and the Exxon Valdez oil spill are associated with sleep deprivation. And in a team that works at the pace of its slowest member, sleepy workers can cause decision-making processes to lag (this is the point at which I should apologise to my Dean for falling asleep at last month’s research committee meeting).

Despite the potential negative impacts of sleep deprivation on those around us, it is probably not all that surprising that policymakers have stayed away from yawn taxation and sleep subsidies. Yet it is a trifle strange that firms do not do more to encourage well-rested employees. While many large Japanese firms have ‘nap rooms’, any Australian who lies down under their desk risks being awoken by the shriek of ‘he’s fainted!’.

We all have a bad night’s sleep from time to time. In certain high-pressure occupations, working late may be the norm. And some people simply have the genetic luck to need less sleep. But for the rest of us, it could be time to start changing the culture. Rather than seeing sleep deprivation as a sign of toughness, perhaps we should begin to regard it as a problem to be addressed. Just as unions fought for the eight-hour day, does Australia need a social movement to bring back the eight-hour night?

Andrew Leigh is a professor in the Research School of Economics at the Australian National University. He is presently a candidate in the ALP preselection for the federal seat of Fraser.

And for those who like decimal points with their data, here are the percentages:

In other words, only 1/10th of those in the poorest decile know it, and only 1/100th of the top decile are willing to admit it.

A decade on, it’d be worth seeing whether Australian are any better informed about their true position in the income distribution than we were in 1999.

Top Earners Getting More Milk From the Cash Cow

Australia’s top earners have increased their share of income more than three fold over the last three decades, according to new research from The Australian National University.

The study, conducted by ANU economist Professor Andrew Leigh from the Research School of Economics, in conjunction with Oxford University’s Sir Tony Atkinson, used taxation statistics to estimate the share of income held by the rich between 1921 and 2007.

The researchers also found that the income share of the country’s best paid workers was rising at a rate far outpacing ordinary workers.

“The income share of the richest one per cent (those earning more than $197,000 in 2007) declined from 1921 to 1980, and has risen since then. In 1980 the top one per cent had five times their share of household income. Now, the same group has 10 times their share of household income.

“This pattern is even starker among the top 0.1 per cent (those earning more than $693,000 in 2007) whose share of household income has more than tripled since 1980,” said Professor Leigh.

He added that the patterns reflected figures about the pay of the country’s CEOs, High Court judges and top public servants.

“From 1993 to 2009, the pay of top-100 CEOs rose twice as fast as the salary of ordinary workers. In that period, the average earnings of CEOs rose by an average of 7.5 per cent per year. Over the same period, average salaries across the economy rose by an average of 3.7 per cent per year.

“In 1993, the average earnings of CEOs in the top 100 Australian firms was about $1 million. By 2009 this had risen to around $3 million.

“The salaries of High Court Judges and the country’s top public servants have also risen faster than average earnings,” he said.

Using newly available taxation statistics and CEO pay data, the study updates results that were first published in 2003. “Several factors drove the increase in inequality. These include internationalisation of labour markets for executives, technological change, and lower marginal tax rates,” he added.

Methodology

· The study combines taxation data with external information (such as population totals, and the total personal income if everyone was required to file a tax return), in order to calculate the share of income held by particular top income groups.

· A potential drawback of this approach is tax avoidance or evasion. For the most part, studies suggest that this is relatively small, and has little impact on trends in top incomes.

· For simplicity, I refer to tax years using their start date (eg. what I call 2007 is the tax year 2007-08).

Top 10 per cent

· In 2007, the cut off for entry into the top 10 per cent was individual income of $73,000.

· In 1941, the top 10 per cent had 34 per cent of all household income. This dropped steadily until 1980, when that group had 25 per cent of household income. Over the past generation, it has risen again, and the top 10 per cent now has 32 percent of household income.

Top 1 per cent

· In 2007, the cut off for entry into the top 1 per cent was individual income of $197,000.

· In 1921, the top 1 per cent had 12 per cent of all household income. Again, this dropped steadily until 1980, when that group had 5 per cent of household income. In 2008, the top 1 per cent had 10 per cent of household income.

· Salary earnings comprised 6/10ths of top 1 per cent incomes in the late-1990s, but only 4/10ths by the late-2000s.

Top 0.1 per cent

· In 2008, the cut off for entry into the top 10 per cent was individual income of $693,000.

· In 1921, the top 0.1 per cent had 4 per cent of all household income. By 1980, that group had just 1 per cent of household income. In 2007, the top 0.1 per cent again had 4 per cent of household income.

CEO Pay

· Over the period 1993-2009, the average earnings of CEOs in the top 100 Australian firms rose by an average of 7.5 per cent per year (Productivity Commission, Executive Remuneration in Australia, p.61). Over the same period, average salaries across the economy rose by an average of 3.7 per cent per year.

· In 1993, the average earnings of CEOs in the top 100 Australian firms was about $1 million. By 2009, this had risen to around $3 million.

Other top groups

· The salaries of High Court Judges and top public servants have followed a similar trajectory.

All data are available in spreadsheet form at http://andrewleigh.org/

Update: The ever-impressive Possum Comitatus takes the spreadsheet and does some fascinating analysis of his own.

Intergenerational Income Mobility in Urban China
This paper estimates the intergenerational income elasticity for urban China, paying careful attention to the potential biases induced by income fluctuations and life cycle effects. Our preferred estimates are that the intergenerational income elasticities are 0.74 for father-son, 0.84 for father-daughter, 0.33 for mother-son, and 0.47 for mother-daughter. This suggests that while China has experienced rapid growth of absolute incomes, the relative position of children in the distribution is largely determined by their parents’ incomes. Investigating possible causal channels, we find that parental education, occupation, and Communist Party membership all play important roles in transmitting economic status from parents to children.

In 2007, New York City’s Center for Economic Opportunity launched Opportunity NYC–Family Rewards, an experimental, privately funded, conditional cash transfer (CCT) program to help families break the cycle of poverty. CCT programs offer cash assistance to reduce immediate hardship, but condition these transfers on families’ efforts to build up their “human capital,” often by developing the education and skills that may reduce their poverty over the longer term. Family Rewards is the first comprehensive CCT program in a developed country.

Aimed at low-income families in six of New York City’s highest-poverty communities, Family Rewards ties cash rewards to pre-specified activities and outcomes in children’s education, families’ preventive health care, and parents’ employment. The three-year program is being operated by Seedco — a private, nonprofit intermediary organization — in partnership with six community-based organizations. It is being evaluated by MDRC through a randomized control trial involving approximately 4,800 families and 11,000 children, half of whom can receive the cash incentives if they meet the required conditions, and half who have been assigned to a control group that cannot receive the incentives. This report presents initial findings during the program’s early operating period.

Key Findings
Despite initial challenges in understanding the program’s large number of incentives and related payment requirements, nearly all families eventually earned rewards — more than $6,000, on average, over the first two years. In addition, effects from Family Rewards varied across a wide range of outcome measures — for example, the program:
? Reduced current poverty and hardship, including hunger and some housing and health care hardships
? Increased savings and the likelihood that parents would have bank accounts, and reduced the use of alternative banking institutions (such as check cashers)
? Did not improve school outcomes overall for elementary or middle school students, but did increase school attendance, course credits, grade advancement, and standardized test results among better-prepared high school students
? Somewhat increased families’ continuous use of health insurance coverage, reduced their reliance on hospital emergency rooms for routine care, and increased their receipt of medical care
? Substantially increased families’ receipt of preventive dental care
? Increased employment in jobs that are not covered by the unemployment insurance (UI) system but reduced employment in UI-covered jobs

Because only the first 12 to 24 months of the program are covered — including a “start-up” phase during which operational “kinks” were being worked out — it is too soon to draw firm conclusions about the full potential of Family Rewards. Future reports will present longer-term findings, eventually covering all three years of program operations plus two additional years after the cash incentives are no longer offered.

Or to put it in context:

The findings indicate that Family Rewards led to small increases in elementary and middle school students’ participation in extracurricular activities and in the extent to which their parents were engaged with their schooling, but it had few effects on attendance or test scores. Among the older students, the program led to large and consistent gains in school progress for a subgroup of more academically prepared ninth-graders. For students in that subgroup, who may have been in a better position to take advantage of the incentives, the program increased attendance and school progression, namely promotion to the tenth grade and number of credits earned. These effects are encouraging and on a par with those found from other, more intensive, school-based interventions

The really interesting question will be whether these impacts are sustained when the cash stops flowing. Nonetheless, the early results are pretty encouraging, and suggest that these kids of innovative programs are at least worth considering in Australia, particularly for Indigenous communities.

Update: The NYT reports that the results have been too modest for NYC to make this a publicly-funded program (HT: AK).

‘High Taxes Not Without Sin’, Australian Financial Review, 30 March 2010

In The Australian Legend, Russell Ward wrote ‘no people on the face of the earth ever absorbed more alcohol per head of population’ than Australians in the 1800s. While scholars debate Ward’s precise claim, it is clear that European settlers consumed vast quantities of alcohol. And 80-90 percent of men smoked.

Substance abuse is inextricably linked to Australian history, where rum and tobacco often took the place of cash. Today, 19 percent of us are regular smokers, while 13 percent of us are risky drinkers (defined as more than 4 drinks a day for men, and more than 2 drinks a day for women). Compared with other developed countries, we have relatively few smokers, and a slightly above-average level of alcohol consumption.

Can sin taxes make us pure? In the case of cigarettes, you might expect that most buyers would be addicts, and therefore unresponsive to prices. But it turns out that smokers are surprisingly price-sensitive. On average, a 10 percent price hike cuts cigarette sales by 5 percent. Although this is offset slightly by an increase in intensity (higher taxes induce smokers to take a few extra puffs out of each cigarette), the health benefits are still substantial. Teens are two to three times as price-responsive as the rest of us.

Put another way, if tax increases were to raise the price of a cigarette from 50 cents to 55 cents, we might expect 150,000 of Australia’s 3 million smokers to kick the habit. The same price rise would probably also deter thousands of high school students from becoming addicted in the first place. Because many smokers want to quit, half of all smokers support higher cigarette taxes – suggesting that such a policy might find favour at the ballot box.

For alcohol, it is less obvious that higher taxes lead to better health outcomes. Unlike smoking, moderate alcohol consumption does not seem to be bad for you (indeed, occasional tipplers may even gain a health benefit). So raising the price of grog is only a good public health measure if it reduces the kind of heavy consumption that is associated with cirrhosis of the liver and misuse (such as drink driving and domestic violence). If higher alcohol taxes deter grandma from having a glass of sherry, but do nothing to prevent the lad on the corner pub from sinking his sixth schooner, we should rate them a public policy failure.

It turns out that alcohol taxes deter both groups. On average, a 10 percent increase in price reduces overall consumption by 5 percent, and lowers heavy drinking by a little less – perhaps around 3 percent. There is also some direct evidence that higher alcohol taxes reduce drink driving, with one US study suggesting that a 10 percent increase in alcohol prices would reduce road fatalities by 6 percent (saving perhaps 90 lives annually). However, because binge drinkers are less price-responsive than the rest of us, alcohol taxes are a blunt instrument for cutting road deaths.

What about the equity implications of sin taxes? In the case of alcohol, drinking rates rise with incomes, so alcohol taxes are modestly progressive.

Cigarette taxes create something of a paradox. Since the poor are more likely to smoke, cigarette taxes are regressive. Yet a rise in cigarette taxes induces more quitters among the low-income population. So for hard-core addicts, cigarette taxes fall more heavily on the poor. But cigarette taxes can be a powerful tool for improving health outcomes among the disadvantaged. Over recent decades, much of the drop in smoking has been due to higher prices, so you don’t have to be a wowser to see the potential for increased taxes to produce better health outcomes.

Will this election give Australian headline writers the chance to bring out those old front pages that said ‘Beer, cigs up’? In conjunction with restrictions on sales and advertising, sin taxes have proved an effective public health policy. Alas, they also have unintended consequences – reducing moderate alcohol consumption, and acting as a regressive tax on low-income smokers who are never going to quit. Even virtuous taxes have their vices.

Andrew Leigh is a professor in the Research School of Economics at the Australian National University. He is presently a candidate in the ALP preselection for the federal seat of Fraser.

In case you’re wondering about whether a 10% rise in alcohol prices is a good way to save 90 lives, here’s my thinking. According to my research assistant Jenny Chesters and the 2003-04 Household Expenditure Survey, Australia spends $9.3 billion per year on alcohol (7.7 million households x $1200), so a 10% price rise is $930 million. Assuming reasonable deadweight costs, it’s likely that the social cost of this exceeds the cost of saving 90 lives (which standard benchmarks would put around $315 million). In other words, for the same societal resources, we could save more lives in other ways.

Of course, the optimal alcohol tax rate would be low on the first drink, and rising steadily thereafter. Fame and riches to anyone who can work out how to implement this in practice.

Other useful sources include this Cato debate. Harry Clarke also has extensive writings on tobacco and alcohol, which I probably should have referenced in the AFR.

In case anyone is interested in the politics of it all, the only publicly available survey I could find that asked about voting and smoking was the 1990 Australian Election Study. On primary votes, never-smokers split 39% ALP/48% Coalition, ex-smokers broke 41% ALP/48% Coalition, and chain smokers (>20/day) voted 52% ALP/35% Coalition.

Is anyone sitting on data that contains both a transcript and the worm level? I’d be curious to see if there are systematic patterns.

Home Computer Use and the Development of Human Capital (gated stable link, ungated unstable link)
Ofer Malamud and Cristian Pop-Eleches
This paper uses a regression discontinuity design to estimate the effect of home computers on child and adolescent outcomes. We collected survey data from households who participated in a unique government program in Romania which allocated vouchers for the purchase of a home computer to low-income children based on a simple ranking of family income. We show that children in households who received a voucher were substantially more likely to own and use a computer than their counterparts who did not receive a voucher. Our main results indicate that that home computer use has both positive and negative effects on the development of human capital. Children who won a voucher had significantly lower school grades in Math, English and Romanian but significantly higher scores in a test of computer skills and in self-reported measures of computer fluency. There is also evidence that winning a voucher increased cognitive ability, as measured by Raven’s Progressive Matrices. We do not find much evidence for an effect on non-cognitive outcomes. Finally, the presence of parental rules regarding computer use and homework appear to mitigate the effects of computer ownership, suggesting that parental monitoring and supervision may be important mediating factors.

~~~~~~~~

The ABS are undertaking the first major review of the Consumer Price Index in 13 years, and Rob Bray (who has just moved into the office next door) has just put a very detailed submission. If you don’t have time to read the entire 58 pages, here’s Rob’s summary:

• The current CPI was specifically selected at the time of the 13th Series Review to be a measure of price change for the purposes of monetary policy:
- In this capacity it less effectively provides a measure of the changes in the Cost of Living for households in Australia, despite the fact that this is the way in which the measure is most frequently perceived.
- Notwithstanding this change, the headline CPI is only one of a wide range of indicators used for monetary policy setting, and the sole focus of the CPI on this hence appears neither justified, and risks the measure not only being inadequate for other purposes, but carries the potential for the measure to lose credibility with the population.

• All price indexes are subject to bias. This is also the case with the CPI. Specifically:
- The use of a fixed basket of goods overstates price change where consumers substitute products and change their consumption pattern;
- The plutocratic nature of the index, which weights the contribution of households to the CPI by their level of expenditure, while making the index representative of price change in spending across the population, means it can be unrepresentative of the spending of ‘average households’; and
- The approach to quality adjustment may mean that the index is not indicative of the actual change in the cost of living.
- As the first of these is likely to bias the index upwards, the second two may have biased it downwards over recent years (as a measure of the change in the cost of living of an ‘average household’); it is unclear what the overall balance of bias is.

• Decisions such as the primary purpose of the CPI are not statistical decisions, but rather involve ABS deciding which balance of interests it will support relative to other users. This policy is then accompanied by an ABS policy to avoid “publishing measures that could be viewed as competing with the headline CPI measure”.
- This approach is poor statistical practice.
- The need to ‘protect the CPI from competition’ hinders an environment of openness and critical review of the CPI.

• Reflecting this the specific issues raised in this submission encompass:
- Weighting:
· Reliance upon fixed weights which do not reflect substitution and changing patterns of consumption between the 6 yearly HES.
· An inappropriate treatment of the costs of home purchasers and owners which means housing costs are underweighted.
· Use of a weighting pattern which is unduly reflective of the consumption patterns of higher income households.

- Prices:
· Whether the ‘constant quality’ adjustments of some items do not realistically reflect either the utility individuals gain from changes in quality not their capacity to meet needs in the market.

- Openness and transparency:
· A less defensive approach to a ‘headline measure’.
· Greater variety of indicators reflecting the differing needs and interests of users.
· Reforms to the approach to revision.
· Further improvements to the amount of information provided on the CPI and its components.
· An active program of research and analysis the results of which are available to users.

Unwilling or Unable to Cheat? Evidence from a Randomized Tax Audit Experiment in Denmark (gated stable link, ungated unstable link)
Henrik Kleven, Martin Knudsen, Claus Kreiner, Soren Pedersen and Emmanuel Saez
This paper analyzes a randomized tax enforcement experiment in Denmark.  In the base year, a stratified and representative sample of over 40,000 individual income tax filers was selected for the experiment.  Half of the tax filers were randomly selected to be thoroughly audited, while the rest were deliberately not audited.  The following year, “threat-of-audit” letters were randomly assigned and sent to tax filers in both groups.    Using comprehensive administrative tax data, we present four main findings.  First, we find that the tax evasion rate is very small (0.3%) for income subject to third-party reporting, but substantial (37%) for self-reported income.  Since 95% of all income is third-party reported, the overall evasion rate is very modest.  Second, using bunching evidence around large and salient kink points of the nonlinear income tax schedule, we find that marginal tax rates have a positive impact on tax evasion, but that this effect is small in comparison to avoidance responses.  Third, we find that prior audits substantially increase self-reported income, implying that individuals update their beliefs about detection probability based on experiencing an audit.    Fourth, threat-of-audit letters also have a significant effect on self-reported income, and the size of this effect depends positively on the audit probability expressed in the letter.  All these empirical results can be explained by extending the standard model of (rational) tax evasion to allow for the key distinction between self-reported and third-party reported incomes.

Update: On a similar theme, here’s Tim Harford on why policymakers should do more randomised trials.

Does Maternal Age Affect Children’s Test Scores?
Andrew Leigh and Xiaodong Gong
We estimate the relationship between maternal age and child outcomes, using indices aimed at measuring overall outcomes, learning outcomes and social outcomes. In all cases,we find evidence that children of older mothers have better outcomes. Not only do children born to mothers in their twenties do better than children born to teen mothers, but children born to mothers in their thirties do better than children born to mothers in their twenties. However, when we control for other socioeconomic characteristics, such as family income, parental education and single parenthood, the coefficients on maternal age become small and statistically insignificant. The only exception is an index of social outcomes, which is positively associated with maternal age, even controlling for socioeconomic factors. For cognitive outcomes, young motherhood appears to be a marker, not a cause, of poor child outcomes.

An easy way to see the result is to compare the relationship between maternal age and child outcomes first without any socioeconomic controls:

(strong positive relationship, mostly statistically significant)

…and then with socioeconomic controls:

(virtually no relationship, generally statistically insignificant).

What will a company tax rise do to prices? While the evidence is thin, theory suggests that companies will be most likely to put up prices on consumers when they do not face competition from importers. So an Australian shoe manufacturer (do we have any left?) may be unable to shift the burden to consumers. But a fast food outlet will have greater capacity to raise prices.

Which prompted this informative response:

Dear Andrew,

After reading your piece in the AFR this morning I thought I’d answer your question: Are there any shoe manufacturers left in Australia? Despite the fact that the footwear industry has undergone massive restructuring, redundancies, factory closures, tariff elimination (despite competitor countries’ excessive rates) and now, almost no government assistance, there is still a solid footwear manufacturing industry in place.

The industrial/safety market within our country is well supplied by local manufacturers such as Oliver & Stevens, Steel Blue, Rossi Boots, Mongel Boots, Redback and Taipan Footwear – these companies continue to make shoes in Australia and supply not only our industrial needs but also critical areas such as the defence forces, fire fighters and police.

In the general consumer area the iconic Australian brand R.M.Williams has an impressive South Australian factory where all their boots are made. In the fashion area companies like J.Robins in Sydney continue to design and manufacture footwear for the women of Australia.

Footwear manufacturing is a very labour intensive industry and, despite the best technology, is a tough business to be in when we have no competitive advantage in the employment of people. However, those manufacturers that remain in Australia have a passion for the industry and a determination to survive.

An interesting resource to find more about the TCF industry is Professor Roy Green’s "Building Innovative Capacity – Review of  the Textile, Clothing and Footwear Industries, 2008"

Regards

Phil Butt
President
Footwear Manufacturers Association of Australia.

‘Abbott Tax Hits Workers’, Australian Financial Review, 16 March 2010

On the morning that Tony Abbott released his proposal to pay for paid parental leave with a tax on Australia’s 3200 largest firms, I was reading Norman Lindsay The Magic Pudding to my three year old son. As you know, it involves a pugilist strolling around outback Australia, punching his enemies on the nose and promising his friends a free lunch. The Magic Pudding has a similar storyline.

Promising to raise company taxes has an visceral appeal to any ambitious opposition. Perhaps some voters will think that they will be borne by the companies themselves, leaving all living persons miraculously unharmed. Slightly savvier citizens might think that company taxes are entirely borne by investors.

A central tenet of public finance, however, is that the entity that has the legal obligation to pay a tax is not necessarily the one that bears the burden. For example, payroll taxes are levied on firms, but we know that they are mostly borne by workers. Raise payroll taxes, and firms cut wages. Lower payroll taxes, and most firms will pass on a pay rise.

The GST is another case in which the burden of a tax doesn’t fall on the entity that pays the tax bill. Although the law says that the tax is levied on those who supply goods and services, it is customers who end up bearing most of the burden.

Which brings us to company taxes. For decades, economists have argued over how the burden of company taxes are shared between investors, employees and customers. In the short-term, it is difficult to change prices and wages, so a higher company tax rate will be paid in the first instance by shareholders.

But over time, the burden is likely to shift. Investors are a footloose bunch, with the ability to shift their money into sectors like real estate where they can avoid company taxes. For an open economy like Australia’s, higher corporate income taxes will lead investors to buy foreign shares instead (which is why small countries have been cutting company tax rates over recent decades). To keep their investors, companies may respond to the tax rise by raising revenue and cutting costs.

What will a company tax rise do to prices? While the evidence is thin, theory suggests that companies will be most likely to put up prices on consumers when they do not face competition from importers. So an Australian shoe manufacturer (do we have any left?) may be unable to shift the burden to consumers. But a fast food outlet will have greater capacity to raise prices.

In the case of wages, the empirical evidence is stronger. In a recent review of the literature, William Gentry (Williams College) concludes that most of the impact of a corporate income tax rise falls on workers. Increase company taxes by 10 percentage points, and wages fall by 6-10 percent.

Assuming Gentry’s finding applies here, this means that Abbott’s paid parental leave plan will be mostly paid for by those who work for big firms. For these companies, Abbott proposes to raise the company tax rate by 1.7 percentage points. If Gentry is right, we should expect to see wage drops of at least 1 percent in these companies.

Which workers will cop the pay cut? Abbott’s plan applies to companies with taxable income over $5 million, and we know that employees of such firms tend to earn a bit more than the average worker. But the difference is largely due to managerial salaries (you can’t be a high-ranking manager in a small firm). Further down the pecking order, there are plenty of modestly-paid workers toiling in the retailers and banks that make up Australia’s largest businesses. If a company tax increase is passed on to employees, these are the people who will pay for parental leave.

Abbott’s plan aside, it is useful to recognize that company taxes are not as progressive as they may appear at first blush. If the corporate income tax was solely an investor tax, it would fall largely on the rich. But if it is mostly a worker tax, then this means that the corporate tax burden is spread broadly across Australian society.

Good economics doesn’t always make good politics. Indeed, politicians can sometimes look as through they’re living by Bill Barnacle’s maxim (‘as we’re perfessional puddin’-owners, we have to fight them on principle’). Enjoy the spectacle, but don’t forget who’s in the pud.

Andrew Leigh is a professor in the Research School of Economics at the Australian National University.

One point that Nicholas made – which I was unable to squeeze in – is that big businesses are already hit with a special tax that falls on workers, because the payroll tax doesn’t apply to small enterprises. Another reason not to be ramping up their company tax rate.

At this point, I wonder if Thai leadership are ruing the eighteenth-century decision to move the seat of government from Thonburi to Bangkok?