Friday, February 27, 2015

Tax Avoidance and the Greek Debt Crisis

With Greece and its debt problems still in the news five and a half years after it began and four years after it reached the critical point in June 2011, the country still has to find a long-term solution to its $374.5 billion potential debt default.  As it stands now, Greece's debt is 168 percent of GDP and it is accruing interest at $28.1 billion annually.

A paper by Nikolaos Artavanis, Adair Morse and Margarita Tsoutsoura at the Booth School of Business at the University of Chicago examines one of the key issues that worsened Greece's fiscal health.  The authors use an interesting approach to estimate income tax evasion from Greece's private sector by examining the amount that banks lend to individuals.  They note that banks are most likely to lend to all individuals based on the banks' perception of true income meaning that tax-evaders will be granted credit far in excess of what they would normally be given if their incomes were as low as they reported to the government.

The authors used household credit data from tens of thousands of applications by wage and self-employed workers for credit products between 2003 and 2010; this data includes credit cards, mortgages, term loans, overdraft facilities, appliance loans and refinancing, with all data sourced  from a large Greek bank, to estimate the true income of individuals.  The bank being used in the study is one of ten large Greek banks that has branches throughout Greece.  Mortgage data, in particular, is useful for determining real household income since individuals that take out a mortgage generally choose to buy as much house as their household income permits and the standard rule of thumb states that mortgage payments should not exceed 30 percent of monthly income.  The authors note that they only had mortgage data from 2006 onward, however, 80 percent of Greek households end up owning a home so the sample size is reasonable.  The authors also use a standard tax evasion assumption which states that for wage workers, reported income is generally equal to real income.  On the other hand, those that are self-employed will find it much easier to declare less income than what they earn since the income paper trail is less robust.

Now, let's look at the findings which vary somewhat depending on the credit facility used (i.e. refinancing, credit cards, mortgages).

From the data on refinancing, lawyers, engineers, accounting, finance and medicine are identified as professions in which the self-employed neglect to declare at least half of their income.  Surprisingly, refinancing data shows that education is a significant tax-evading industry.  This is because many families in Greece hire private tutors for their school-aged children.  As well, those who are employed in media, particularly journalists, are high tax-evaders, largely because they have influence over political decision-making and have enjoyed relatively little scrutiny regarding their incomes.

From the data on credit cards, the authors looked at credit card limits with the largest being 35,000 euros.  This data shows that the biggest tax evaders are employed in education, construction, law, the media and the arts.  Accounting, financial services and medicine also appear, however, at rates that are slightly lower than for other types of credit since the credit card model is less robust for high income individuals.

From the data on mortgage payments, the authors found that accountants, financial service professionals, doctors and engineers are the biggest tax evaders.  Lawyers have slightly lower tax evasion that what is shown for other types of credit but still appear high on the list of income under-reporting.

The authors found that, in general, tax evasion increased as wealth increased as shown on this graph which shows the difference between reported income and real income for each 5th of a percentile:


Overall, the biggest reported-to-true income multipliers by industry in Greece are as follows:

1.) Education.

2.) Medicine.

3.) Engineering.

4.) Law.

5.) Media.

6.) Fabrication.

7.) Accounting and Financial Services.

On average, these occupations reported well less than half of what is actually earned.  It is interesting to see that these occupations require advanced degrees and that revenue depends on reputation.

In terms of the amount of tax evaded as measured in euros, the ranking is as follows:

1.) Doctors.

2.) Private Tutors.

3.) Engineers.

4.) Lawyers.

5.) Accounting and Financial Services

Here is a table showing the level of tax evasion in euros by occupation:


As an aside, Transparency International's National Survey on Corruption in Greece for 2010 showed that Greeks reported paying the largest bribes to hospitals, followed by lawyers, doctors, banks, vehicle inspection centers, clinics, civil engineers and other engineers.  The payment of bribes is the most significant way that wage earners can evade taxes.  

For the sake of completeness, here is a map showing the percentage of tax evasion by Greek zip code with the darker colours showing a higher level of tax evasion:


The red circled area is Larissa which, coincidentally, has the largest number of Porsche Cayennes in Europe.


One of the biggest problems facing Greece is its inability to collect income taxes owed by some of the nation's wealthiest and most highly educated professionals.  Using the data in this study, the authors estimate that, in 2009 alone, 28 billion euros in self-employed income went untaxed which amounted to 31 percent of the government's deficit in 2009 or 48 percent of the deficit in 2008.  Without more a more robust tax collecting mechanism, all of the actions by the EU, IMF and World Bank will obviously not solve Greece's long-term debt problems.

Thursday, February 26, 2015

What is the Wage - Inflation Relationship Telling Us?

There are many employment statistics that play second fiddle to the headline U-3 data but two of them, the employment cost index and average hourly earnings, affect every American that still has a job.  

Let's start by looking at the employment cost index (ECI).  This statistic measures the cost of workers' wages and salaries as well as benefits and bonuses at all levels in Corporate America.  The Bureau of Labor Statistics bases the ECI on a survey of employer payrolls in the final month of each quarter.  The ECI is considered a leading indicator of inflation since compensation generally increases prior to companies increasing prices for its goods and services.

Here is a graph from FRED showing the employment cost index since 2001:


While it might appear to be a steady increase, such is not the case.  Here is the same data showing the percentage change from the previous year:


Since the end of the Great Recession, it is quite apparent that growth in the employment cost index has been very low, ranging from 1.4 percent in Q3 2009 to 2.3 percent in Q4 2014 and has grown at an average rate of 1.83 percent over the five year period.  This compares to an average increase of 3.0 percent annually over the period from Q1 2002 to Q1 2008.  While there has been upward pressure in employee costs over the last two quarters of 2014, the data would suggest that there is very little upward pressure on inflation related to increases in compensation.

Now, let's look at average hourly earnings.  Here is a graph from FRED showing average hourly earnings of all employees in the private sector since  early 2006:


Once again, while it may appear to be a slow and steady increase, the same data showing the percent change from the previous year would suggest otherwise:

In December 2014, average hourly earnings had increased from the previous year by only 1.65 percent.  Since the end of the Great Recession, on a year-over-year basis, average hourly earnings have increased by an average of 2.0 percent.  While the data prior to the Great Recession is not complete, on a year-over-year basis, average hourly earnings increased by an average of 3.4 percent.  

To get a better sense of pre-Great Recession earnings growth, while it doesn't cover all non-farm occupations, we have a more complete database when we look at year-over-year increases in average hourly earnings of production and non-supervisory employees as shown here:


It is quite apparent that the Great Recession had a significant impact on wage growth.  During the years between the 1990 - 1991 and 2001 recessions, the average wage grew by 3.3 percent on a year-over-year basis.  During the years between the 2001 and 2008 recessions, the average wage grew by 3.1 percent on a year-over-year basis.  This is substantially more than the 2.0 percent average annual increase since the end of the Great Recession.

The Federal Reserve is obviously watching wage growth since increased wages foretell inflationary pressures.  Back in December 2006 when Alan Greenspan was her boss, Janet Yellen expressed concerns that increasing employee compensation could have put increasing upward pressure on inflation as shown in this comment:

"On the one hand, recent labor market data point to a lower path for the unemployment rate than before, and all else being equal, this boosts our inflation forecast a bit. Offsetting this effect, on the other hand, is the huge downward revision in compensation per hour. When these data came out, I let out a big sigh of relief. The revised data are more consistent with the indications we were getting from the employment cost index and suggest that wage growth has remained contained."


Ms. Yellen made it quite clear that her priority was to control inflation and that she was quite pleased to see a downward revision in employee hourly compensation.  From both the employment cost index and the average hourly earnings data, it looks like Ms. Yellen and her merry band of central bankers will be able to sleep comfortably at night, knowing that the modest increases in compensation for America's workers will put little upward pressure on inflation.  That's one less thing that she needs to worry about when contemplating when she should push interest rates up.


Tuesday, February 24, 2015

United States Debt By President

Updated March 2016

I haven't posted on the current debt situation in the United States for some time so I thought that it was time to revisit my debt-by-President information to give us a sense of how the United States became the world's largest debtor state.

For the purposes of this posting, I source the data from the TreasuryDirect website which provides the public with a year-by-year and month-by-month update of the debt position and debt activity.  I have gone back to the Kennedy Administration and have used the debt figures for the end of the month of January during each inauguration as a starting and ending point, except in the cases of the Kennedy, Johnson and Nixon Administrations when I used the debt data from the end of their last month of service.  In all cases I have included both the external and intergovernmental debt in my calculations

I have made three calculations:

1.) the nominal growth of the federal debt during each President's tenure in the Oval Office.

2.) the percentage growth of the debt from the end of the month of each President's inauguration to the end of their final month in the Oval Office.

3.) the compound annual growth rate (CAGR) of the federal debt during each President's tenure in the Oval Office which is calculated using this formula:


CAGR provides us with a measure that shows us the compounded average annual growth rate of the debt (in this case) over  all of the years that each President served .

Let's look at a graph from FRED which shows what has happened to the federal debt as a percentage of GDP since 1966:


As you can see, since the fourth quarter of 2012, for the most part, the federal debt has been in excess of 100 percent of GDP, up from 62.8 percent in the fourth quarter of 2007 when the Great Recession was just beginning.

Now, let's look at some graphs that show what has happened to the federal debt under each of the last ten administrations, starting with a graph that shows the federal debt at the end of each President's term (excluding Barak Obama who still has nearly one year left in his second term):


It is rather interesting to note how the growth in the nominal level of the debt has accelerated, particularly since the Bush II Administration. 

Here is a bar graph which shows the nominal increase in the federal debt for each of the last ten Administrations:


In nominal terms the debt has grown (and will continue to grow) by the most under the current President.

Here is a bar graph which shows the percentage increase in the federal debt for each of the last ten Administrations:


By a very wide margin, the percentage growth in the federal debt was the highest under the Reagan Administration with the debt growing by 188.84 percent.  Under the Bush II Administration, the debt grew by 86 percent and under the current Obama Administration, the debt has grown by 79.6 percent, however, this will rise over the next two years of Barak Obama's presidency.

Lastly, here is a bar graph showing the compounded annual growth rate of the debt under each of the last ten Administrations:


Again, the Reagan Administration comes in first place with an annual debt growth rate of 14.18 percent, followed by the Ford Administration at 12.99 percent and the Bush I Administration at 11.48 percent.


Let's close by looking at one last statistic.  If we take the current federal debt of $19.093 trillion and divide it by 318,857,056 Americans, the per capita debt share of every American man, woman and child is:

                                              $59,853 

To put this number into perspective, according to the 2013 American Community Survey, the median household income in the United States in 2013 was $52,250.