Wednesday, October 26, 2022

Some notes to my IB Economics students on the MULTIPLIER EFFECT

 THE MULTIPLIER EFFECT

A (Keynesian) idea whereby an increase in government expenditures G will lead to a greater increase in national income (in real GDP)

i.e. ΔY > ΔG or, ΔY = κ ΔG

(where little κ is the multiplier)

Let the government increase government expenditures by $100mil. If national income increases by $300mil, the multiplier (little κ) is ‘3’: $300m/ = 3 x $100m

 Why should we expect this to be the case? Because:

  • Your spending is my income.
  • Economic activity takes place in successive rounds (as clearly illustrated by the ‘circular flow of income’ diagram; use in an essay if you consider it necessary)

If the government spends $100 to build a tiny road in my neighborhood and I’m the only worker / engineer, it means that my income increases by $100.

Part of this additional income earned, I will spend on domestic goods and services, say $80. For example, from the additional $100 I earned, I will spend $80 for a dancing company to come at my birthday party and dance for my guests. National income will have increased by $100 + $80 = $180 (and note that two ‘goods’ will have been produced: a tiny road and a dancing performance).

Generalizing: If out of each dollar earned as additional income people spend 80 cents then the so-called Marginal Propensity to Consume (MPC) is 0.80

MPC = ΔC / ΔY, equal to say ‘b’

Let the dancers now spend 0.8 of their additional income on violin lessons for their offspring.  They thus spend $64 (0.8x80) on the violin teachers.  Violin teachers now have additional income equal to $64 and national income has increased by a total of $100+$80+$64 = $244.  They in turn will spend part (0.8) of their additional income on feta cheese…and so on and so forth…

More generally (a geometric series; sum of; converges as 0 < b < 1):

ΔY = ΔG + bΔG + b^2ΔG + b^3ΔG + …

ΔY = (1+b+b^2 +b^3 +b^4 +…) x ΔG

ΔY = (1/(1-b)) x ΔG, so little κ:

κ = {1 / (1-MPCd)] = {1/(MPW)], where MPW = MPS + MPT + MPM

(as if I spend out of $100 of additional income on domestic goods, the rest (i.e. $20) I have either saved (S) or paid in taxes (T) or spent, but on imports M): all three (S,T, M are withdrawals / leakages from the circular flow)

(MPS is the marginal propensity to save, MPT is the marginal tax rate and MPM is the marginal propensity to spend on imports: their sum [MPS+MPT+MPM] is the marginal propensity to withdraw : So (1-MPCd) = MPW (remember the circular flow again)  

Note:

  • The multiplier effect holds if any of the 3 injections increases (remember that injections 'J' include government expenditures G, but also exports X and private investment I) (note that the export multiplier is responsible for the international transmission of the business cycle)
  • The importance of the Keynesian multiplier of government expenditures (G) is that it shows that expansionary fiscal policy in the form of increased government expenditures (G) is very powerful to help an economy grow - lift an economy out of a recession (but, remember the other side, the crowding-out effect)
  • Expansionary fiscal is not only a result of an increase in G but could also be a result of a decrease in direct taxes T.

But, the multiplier effect of an equal (to ΔG) decrease in taxes T is smaller. Why?

Because the first-round effect does not exist. When a government decreases my taxes by $100, my income does not rise by $100: it is my disposable income that rises by $100 so the multiplier process starts with the 2nd round effect of an equal increase in G (since I will spend $80 of this increase in my disposable income on the dancing company above). So, the tax multiplier will be [b/(1-b)]  (the first term of the geometric series is now ‘b’, not ‘1’) It follows that if b is 0.8, then the government expenditure multiplier will be 1 / 0.2 = 5 but the tax multiplier will be 0.8 / 0.2 = 4 (smaller)

  • The multiplier operates also in the opposite direction. A decrease in government expenditures (G) will lead to a greater decrease in national income (Y). When Greece was ‘forced’ to borrow to avoid default during the Greek debt crisis, it ‘agreed’ to decrease government expenditures (G) as part of the agreement.  The IMF had then estimated that the Greek government expenditure multiplier, was, say equal to  ‘2’, so that a decrease in Greek government expenditures (G) by, say, €200mil., was expected to lead to a decrease of Greece’s national income by €400mil. (2x200)

BUT the recession that ensued (the decrease in national income) proved much more severe than expected: Greece’s rGDP decreased by much more than estimated by the IMF, say, by €500 million, and consequently even more people lost their jobs and their income, forcing the IMF (and its then Chief Economist, Olivier Blanchard - now at PIIE), to issue what was referred to as a “mea culpa”: a new IMF paper re-estimated ‘correctly’ the Greek little ‘κ’.

[Please check the Study Guide for other Real World Examples of the multiplier for use in part (b) of a Paper 1 essay]

Endnote:

In Keynes' General Theory we read (p. 128, Palgrave 2007 edition):

"If the Treasury were to fill old bottles with banknotes, bury them at suitable depths in disused coal mines which are then filled up to the surface with town rubbish, and leave to the private enterprise...to dig the notes up again...there need be no more unemployment and, with the help of the repercussions, the real income of the community...would probably become a good deal greater than it is."

Friday, October 14, 2022

National (Public) Debt - some notes: perhaps useful for IB Economics students

 Another 'new kid on the block' for IB Economics students (new syllabus)

What is the national (public) debt?

It refers to what a government owes to all of its lenders: ‘Public debt is created through government borrowing from individuals, corporations, institutions, and other governments’ [from the New Palgrave Dictionary of Economics]

è Debt is created when a government records a ‘budget deficit’: when G > T (when government expenditures exceed government income which is mostly tax revenues (T) but may also be any one-off proceeds from privatizations (the transfer [=sale] of state-owned assets to the private sector) and any profits from state-owned enterprises)

è It follows that the public debt is the sum of all past budget deficits minus any budget surpluses [‘the public debt a stock at any given point in time and represents the net accumulation of the associated deficits over all previous time periods’; from the New Palgrave]

è Given that a government borrows by issuing bonds  (=selling to whoever is willing to buy these i.e. to lenders) it follows also that the debt is equal to the value of all outstanding bonds (i.e. bonds that are held by its creditors and thus have not yet ‘matured’ and been paid off)

è Government bonds may be held by domestic creditors (=lenders) or by foreigners (= external debt; denominated typically but not necessarily in USD); bonds are held by individuals, firms, banks, universities, pension funds, money managers etc. etc.

To compare the size of the debt of a country with other countries or to determine the extent to which it is ‘sustainable’ , the absolute ‘dollar’ size of the debt is expressed as a proportion of GDP (scaling for the size of the economy).  We thus focus on the debt to GDP ratio.

(perhaps subscribe for free to Finance & Development, an IMF publication; the IMF will send you at your address every quarter the newest issue: https://www.imf.org/en/Publications/fandd

Perhaps read read: (a) https://www.imf.org/en/Publications/fandd/issues/2022/03/Balancing-act-Gaspar (b) https://www.imf.org/en/Publications/fandd/issues/2022/03/Deciding-when-debt-becomes-unsafe-Blanchard

Perhaps also check out the post on Jason Furman here: http://www.ibeconomics.org/2021/09/the-national-debt-expressed-as.html

Note:

A more appropriate metric of debt is the ‘debt service over GDP’.  The reason is that debt is a ‘stock’ variable (measured at a point in time, (say, December 31st of the year) whereas GDP is a ‘flow variable’ (measured over a time period, typically over a year).  If Janice earns income of $100000 per year and has borrowed to buy a house $300000 dollars (so owes banks $300000 (her debt)),  then her debt to her income ratio is 300000/100000 or 300% of her income - but if she needs $10 000 per year to service her debt then the debt service to her income ratio is 20% and what matters to Janice is what she pays off annually to her creditor (the bank).  That is why the debt service to GDP ratio is more informative than the simple debt to GDP ratio.

Cost of high debt 

Cost of debt servicing: 

-      a. Opportunity cost of the funds sacrificed as these funds could have been invested by the government in pro-growth and pro-development goals i.e. on infrastructure investments, on education and on health care services; debt servicing thus restricts the ‘discretionary spending’ ability of the government

-      b. Policymakers cannot as easily adopt expansionary fiscal policy (G up; and/or T down) since It would lead to more debt accumulating

-      c. If debt is mostly domestically held (as it is in Japan) and if the country (unlike Greece- a member of the Eurozone) can issue its own currency, then, if debt is getting dangerously high, the government may be tempted to ‘monetize’ its debt i.e. to print more money to pay off its creditor; but then there is higher risk of inflation (remember Friedman’s helicopter money drop: ‘too much money chasing after too few goods’); on the other hand, if an attempt is made to pay off foreign lenders (in USD) by printing more domestic currency then the currency will massively depreciate also creating higher inflation as prices of imported goods will rise.

·        Impact on credit rating:

Credit rating agencies (CRAs) measure the credit risk of debt for all types of investors. 'Their measurement of credit risk includes default probabilities…and serve an economic purpose as they reduce asymmetric information about issuers that investors face when making investments’(from the New Palgrave). Three agencies (Moody’s, Finch, Standard & Poor’s: the ‘big three’) rate the debt of countries and assign a ‘grade’.  Top rated debt (bonds) is AAA which means ultra-safe: creditors will definitely get their money back. But if these agencies come to a realization that a country’s debt is becoming riskier then it will downgrade it. But then lenders will demand a ‘risk premium’ i.e. to earn higher interest rates. This automatically makes the debt burden heavier so a ‘feedback loop’ may be initiated that may lead to default  [and here comes the IMF with its rescue packages (conditional lending at preferential interest rates) that (once 'agreed' upon) hopefully signal to private investors that the country is back on track to become creditworthy].

·        Impact on future taxation and government spending:

This is where ‘austerity policies’ come into play: the debtor country is forced (‘agrees’= more like some ‘arm twisting’) to slash G and to raise T in order to achieve ‘primary budget surpluses’ (a primary budget surplus is recorded when government spending excluding debt servicing is less than government income (T) so that the government can start lowering its debt; Greece was required for a number of years to achieve primary budget surpluses equal to 3% of its GDP which were huge and inflicted significant pain on mostly low income Greek households

-      Note an influential paper by Alberto Alesina and Silvia Ardagna that tried to show that fiscal contraction will not have the expected Keynesian result (to lead to recession) but will lead to expansion (referred to as ‘expansionary fiscal contraction’); Why? Because according to A&A the private sector would see that the government is rationalizing (cutting) its spending that would result in greater confidence levels and thus more investment and consumption from the private sector. Proved to not really be the case. Krugman called this the ‘confidence fairy’)(see https://www.bruegel.org/blog-post/expansionary-fiscal-contractions-and-uk-experiment)

-      Also note paper by Kenneth Rogoff and Carmen Reihhart of Harvard that showed that when the debt to GDP level exceeds 90% then an economy will fall into recession. A graduate student at UMass @Amherst trying as an exercise in econometrics to duplicate their results couldn’t. He and his advisor found that R&R had (a) excluded from their sample countries which if they were included would NOT have led to recession but to continued growth and (b) that there were coding ‘errors’ in their excel files. Paper was largely discredited but it had already damaged many in debt-ridden economies.

-      The ‘mea culpa’ of the IMF:  The IMF had underestimated the Greek multiplier so that the agreed upon cut in Government expenditures  by the Greek government led to a much bigger than projected decrease in national income (which meant that many more than expected suffered a loss in jobs/ income). Olivier Blanchard, then Chief Economist at the Fund (now at PIIE), was forced to issue a correction of the multiplier which was referred to as Blanchard’s ‘mea culpa’.

 

Monday, October 10, 2022

The notes to my IB economics (HL) on the Phillips Curve

For whatever they're worth I am posting here the notes that I have prepared for my students on the Phillips Curve.  

The Phillips Curve (notes by CZ - The IB at Athens College [901])

1958:  A.W. Phillips (LSE) published an empirical paper:

He collected annual UK data on the rate of unemployment and the percentage change in money wages for 96 years: 1861 – 1957 (money wage: whatever is written on your paycheck)

He noticed an inverse relationship between the annual percentage change in money wages and the annual UK unemployment rate: years when UK unemployment was low (= ‘tight’ labor), money wages increased significantly whereas years when unemployment was high (=’slack’ labor market), money wages increased by a little or not at all (compared to the previous year) or may even had decreased.

BUT:

·        If the wage bill for firms represents a big chunk of their total production costs

·        And, if prices are set by firms as a ‘markup’ on unit costs (i.e. say, 10% or 50% more than the average cost of producing, say, ‘pencils’)

THEN, if money wages rise a lot, prices will rise a lot (i.e. inflation accelerates) and if money wages increase a little (or, decrease) then inflation will be lower (i.e. disinflation)

Thinking along these lines many economists (Samuelson, Lipsey et al.) checked for many countries & for different time periods the data on the behavior between the annual rate of inflation and the rate of unemployment. And, OMG, they found that in many countries over many periods of time an inverse (=negative) relationship existed between inflation and unemployment: if unemployment was decreasing, inflation was increasing and if inflation was decreasing, unemployment was increasing.

This inverse relationship between the annual rate of inflation and the rate of unemployment is referred to as the (original) Phillips curve.


This empirical relationship was fully compatible with the ruling at the time Keynesian theory (remember that in the Keynesian model the level of equilibrium real output - of economic activity - is 'demand-driven': it is ‘effective’ (aggregate for us) demand that determines the level of economic activity [the equilibrium real output]):


If AD rises (from AD1 to AD2), then real GDP rises (from Y1 to Y2) so unemployment falls but inflation increases (APL1 to APL2); if AD decreases (from AD1 to AD3) then unemployment rises but inflation decreases (APL1 to APL3).

The original PC illustrated a trade-off between inflation and unemployment and if this relationship is stable through time for a country, then it was as if it presented policymakers a ‘menu of choices’: they could choose that combination on the country’s Phillips Curve that was considered the most desirable.

They could choose and achieve the desired combination using demand-side policies (fiscal (i.e. ΔG, ΔT) and monetary (i.e. Δr).

They even thought back then that they could ‘fine-tune’ the economy (which was of course considered by many as ‘hubris’).

BUT, in the early 1970s this ‘stable’ relationship between the rate of inflation and the rate of unemployment of a country collapsed.

Economies started witnessing something that was incompatible with the Keynesian model: BOTH the rate of inflation AND the rate of unemployment were increasing!  It was as if the Phillips Curve was shifting outwards (to the right).

This phenomenon was referred to as stagflation (=stagnation + inflation).

Arthur Okun (Okun’s Law: a 1% decrease in real GNP growth was associated with a 0.3% increase in unemployment) back then devised his Economic Discomfort Index — which Ronald Reagan renamed the Misery Index — the sum of the unemployment rate and the inflation rate. 

This collapse was the result of the 1st oil crisis, as in 1973, OPEC to ‘punish’ the West  for supporting Israel, restricted the supply of oil and thus quadrupled overnight the price of a barrel of oil (think now of the price of natural gas surging given that Russia is restricting natural gas exports and its impact on European and other economies).

We now realize (with our  current AD and AS tools) that because production costs increased across the board, the SRAS decreased and shifted left leading to higher inflation AND higher unemployment.

MOST IMPORTANTLY, in 1968, Milton Friedman (the most renowned Monetarist; Nobel prize) published a most influential paper titled “The Role of Monetary Policy” where he introduced the term Natural Rate of Unemployment (NRU).

His analysis of the Phillips Curve relationship distinguished between the ‘short run’ and the ‘long run’ (Remember: short run when only some, but not all adjustments are possible; long run is when all adjustments are possible)

·        He claimed that if there is a trade-off between the 2 variables it exists ONLY in the short run; in the long run, there is only one rate of unemployment, the natural rate, which is compatible with ANY rate of inflation as long as this rate of inflation does not change (does not accelerate).  Thus, in the long run the Phillips Curve according to Friedman is VERTICAL at the NRU.

·        He included in his analysis ‘expectations’ about next year’s inflation that workers form à the ‘expectations-augmented Phillips Curve’ or, the “Phelps-Friedman Critique’ {Ed Phelps came up with pretty much the same analysis independently}

·        Friedman claimed that workers form their expectations ‘adaptively’: they form their expectations about next year’s inflation by looking at last year’s inflation (‘backward’ looking expectations) (really, a weighted average of previous rates)

·        Workers thus suffer from “money illusion” i.e. they do not realize immediately that their real wage (the purchasing power of their money wage) decreased when inflation accelerated à their expectations of inflation are slow to adjust (‘adaptive’)

·        If the government tries to lower unemployment below the NRU using expansionary fiscal (and perhaps easy monetary), inflation will accelerate and workers will not immediately realize that inflation is higher than expected and thus their real wage has decreased: they will thus accept jobs offered by businesses who face lower real costs à unemployment does fall but inflation is now higher.


 When expectations of inflation adjust (long run) so that expected inflation equals actual inflation, workers will demand higher money wages (remember that money wages are flexible fully adjust to changes in the APL in the monetarist model) until the real wage is ‘restored’ back to its original equilibrium level so that in the long run, unemployment will return to its natural rate BUT with higher inflation (this analysis is the ‘mirror image’ of the AD/SRAS/LRAS model explaining how an inflationary gap is closed in the Monetarist model – see the SG)

·        To maintain thus unemployment below the NRU with expansionary demand side policies a government would have to engineer ever accelerating inflation, which of course does not make much sense.

·        Friedman’s policy recommendation is thus: “Ms. Policymaker, do not try to lower unemployment below the NRU using expansionary policies; if you are successful, your success only will be short-lived (temporary) because workers’ expectations of inflation will [eventually] adjust and thus unemployment will return to its NRU but with higher inflation”

·        Friedman quoted Abraham Lincoln that ‘you can fool some of the people all of the time, all of the people some of the time BUT NOT all of the people, all of the time”. 

      Remember:  Short run: when actual inflation exceeds expected inflation 

         Long run: when expectations of inflation have adjusted so that expected inflation is equal to actual inflation. 

       Note that now in a part (a) essay asking to explain the NRU we now have three points to explain:

1. It is the unemployment that exists when the economy is at its potential level of real output

2. It is the unemployment that exists when the labor market is in equilibrium (see in the OUP Study Guide p. 92 the Dornbusch & Fisher LD/AJ/LF labor market diagram with the real wage on the vertical) 

3. From the above analysis we now realize that the NRU is the lowest unemployment that can be achieved without inflation accelerating (NAIRU=non-accelerating inflation rate of unemployment)

Also note that unemployment in the US after the 2009 crisis was continuously decreasing: it reached 5%, then 4.5%, then 4.3%, then 4.0% then 3.8% without inflation accelerating.

So many economists as unemployment was decreasing were afraid that if the Fed did not hit soon enough the brakes (i.e. start tightening monetary policy i.e. raising ‘r’), inflation would start showing its ugly face.

Janet Yellen, who was then the Chair of the Fed, resisted these calls and kept interest rates very low (continuing ‘quantitative easing’) à unemployment thus continued to fall, reaching a 50-year record low at 3.5% WITHOUT inflation exceeding the 2% target.[BTW, Yellen's successor, Ben Bernanke, today shared with two other economists the 2022 Nobel Prize in Economics]

Thus, thousands of American households found a job à if the Fed had tightened earlier, as several economists and politicians were insisting, many poor US households would have remained unemployed (on the other hand, quantitative easing [QE] increased income inequality as explained elsewhere)

Back then, many economists had started wondering whether the PC is ‘dead’, whether it was ‘hibernating’, why has it ‘flattened’ etc.

Explanations for the observed phenomenon varied. A couple of accessible to IB Higher Level Economics students include: 

a.   Discouraged workers re-entering the labor market and finding jobs so that U, the numerator, in the unemployment statistic didn’t rise but the labor force number (the denominator) increased, decreasing the rate of unemployment {w/o need for firms to raise wages and thus prices to attract workers}.

b.   Expectations of inflation back then were well ‘anchored’ at the 2% target rate so that neither firms felt the need to raise prices, nor did workers feel the need to demand higher money wages: all stakeholders expected inflation to remain at 2%.

This explains why NOW, the Fed and analysts fear that inflationary expectations are ‘unmoored’ (=de-anchored) which, if true, implies that US inflation will continue to rise. A Central Bank must be credible – which is why Powell (the Chair of the US Fed) sounds so firm about the Fed’s intentions.  [many claim and they seem to be right that the Fed should have tightened monetary policy earlier - it was 'behind the curve']

PLEASE watch Aspen’s Roundtable on the US Economy and pay close attention to what exactly Neel Kashkari (President of the Federal Reserve Bank of Minneapolis) says: 

http://www.ibeconomics.org/2022/08/a-most-interesting-roundtable-organized.html