"The trade-offs (between growth and inflation) are getting less attractive at this point."

That was Bernanke during his April 27th press conference. What's happened since then?

Even if Bernanke was correct then, it seems clear that the trade-offs are once again incredibly attractive.

Billion Prices Project vs QE2

The Billion Prices Project uses online data to measure prices on a daily basis. Their results  closely follow non-seasonally adjusted CPI.

You can see the important influence commodity prices have had on CPI, but seeing as how those were, at least partially, a result of QE2* and still don't fully explain the rise in prices after QE2 I see this graph as another strong piece of evidence that QE2 worked.

I expect price increases to continue to moderate in the future. The economy still can benefit greatly from more aggregate demand.

*Commodity price increases as a result of higher world aggregate demand (QE2) are reflecting something good, price increases as a result of supply shocks (Libya) are bad.  This is economics 101 but it is still very often misunderstood.

QE2 and the Economy

QE2 clearly impacted asset prices, but how was the actual economy affected? 

It should be noted that the recovery had basically lost all momentum in the summer of 2010 and that claims that the economy was "already on the road to recovery" before QE2 seem questionable, especially given the impact QE2 seemed to have on asset markets.

Starting with employment:

\

Employment based on the Establishment Survey (the survey used to measure month to month changes in employment) clearly did better during the "QE2 period" than during the period before QE2. If employment growth continues at a 150,000+ pace over the next couple months this may just be momentum but if it remains weak (and I predict it will) it seems pretty clear QE2 positively affected employment growth.

The Household Survey (The survey used to measure unemployment) basically gives the same result. Simply put, the only time we've had adequate employment growth since the recession was during the period of QE2.

Since unemployment claims are measured on a weekly basis they provide a somewhat clearer picture of the impact of QE2 on employment. Unemployment claims sharply refute the claim that things were improving without QE2 (notice the "flatness" of claims from late 2009 to mid 2010) or that QE2 didn't help stimulate aggregate demand. This, along with the other employment data seem to strongly suggest QE2 positively affected the employment situation.

It's much harder to argue the turnaround in consumer credit was the continuation of the recovery before QE2. Consumer credit clearly fell until QE2 was hinted at and it started rising steadily immediately afterwards.

Both ISM surveys give the same basic result; the economy slowly gained momentum from late 2008 to early 2010 when it began to stall until QE2 began.

Industrial production is somewhat less clear, but it too seems to have slowed to a crawl in 2010 until a few months after QE2 began. It should once again be noted that economic expectations fell off a cliff before QE2 (very similar to the way they have today), so even a continuation of the previous trend is a success of some kind.

At no point since the recession did retail and auto sales grow so consistently and so strongly than they did during the "QE2 period". Also note the drop-off since QE2 effectively ended.

It will be interesting to see how data turns out the next few months. The weaker it is, the stronger the evidence that QE2 was effective. 

In summary : The evidence that QE2 worked is enormous, but it wasn't enough. If the fed engages in further asset purchases, it should include an explicit target (inflation, NGDP) and should adjust their size according to changes in the performance of the economy.

How QE2's Announcement and Premature End Affected Asset Prices

Firstly, let me justify the beginning and end dates for QE2.

I consider August 27, 2010 to be the effective beginning of QE2. On this day Bernanke delivered the Jackson Hole speech in which he first hinted at another round of quantitative easing for the purpose of stimulating aggregate demand.

I consider April 27, 2011 to be the effective end of QE2 (although it would be more accurate to say "the end of any chance for additional easing of any kind unless things get much much worse"). The key quote by Bernanke was this :

“The trade-offs are getting — are getting less attractive at this point. Inflation has gotten higher. Inflation expectations are a bit higher. It’s not clear that we can get substantial improvements in payrolls without some additional inflation risk."

He hasn't given any indication he has changed his mind since, or that he plans to pursue any other form of easing.

Now that you know where I'm coming from, let's look at some asset prices!

No form of assets responded more clearly to the announcement and effective end of QE2 than equities did. Higher equity prices reflect better economic expectations and those expectations can become a self fulfilling prophecy through Tobin's q and wealth effects. Perhaps more importantly, higher stock prices also partially reflect a lower demand for dollars and that decreased demand immediately increases nominal spending. Also note the fall after QE2's "end".

Similarly, inflation expectations seem to have been strongly influenced by QE2's announcement. Again note the fall since QE2's end.

Treasury yields also noticeably rose in response to QE2. Many have pointed out that QE2 was supposed to reduce yields and have pointed to their rise as evidence against QE2's effectiveness (ignoring equity prices and inflation expectations), but higher yields were actually reflecting higher inflation real growth expectations. As Milton Friedman and Frederick Mishkin have pointed out, interest rates are NOT a reliable indicator of monetary policy and low interest rates can (and do) signal that money is tight. Higher interest rates usually reflect a healthier economy when rates are this low. Unsurprisingly, the lower rates of the past month have been associated with worsening economic conditions.

Finally, the dollar has almost certainly fallen as a result to QE2. A falling dollar can be a bad sign if it is associated with high inflation (or supply side issues), but high inflation is still not a concern. In this case, a falling dollar helped increase net exports (by cheapening domestic goods) and as reflected less dollar hoarding (which is associated with lower V and more NGDP at any given supply of money). And once again, just as any prospects for additional easing ended the dollar began to rise.

Several excellent economists including John Cochran and James Hamilton have expressed doubts about the effectiveness of QE2. It's true that from a purely mechanical perspective, QE2 was likely irrelevant. So why the impact on markets?

The most likely explanation is that QE2 impacted medium term (2-10 years in the future) NGDP expectations. Anything the Fed can do to convince people that they will push for a higher NGDP in the future will improve expectations. Improved expectations have immediate effects on NGDP as they raise expected inflation/real growth and reduce the demand for dollars (thereby increasing V) today.

If this is the case, why does the Fed seem so uncomfortable with higher future NGDP? I think it's mostly a year over year % change issue versus a level targeting one. The Fed is worried that even temporarily higher rates of inflation and NGDP growth (to catch up to past trends) will be hard to push back down. That would be reasonable if expected inflation was above 3%, but at barely 2% these concerns are completely unjustified. The Fed is also likely concerned about bubbles, but with unemployment at 9% and a tremendous amount of slack in the economy, those concerns need much more justification and evidence backing them before they can be legitimately used to prevent easier policy.

(Half way through creating this post I saw that Marcus Nunes beat me to it and posted something very similar. He makes somewhat different -- although equally valid -- points)

In the next few days I'll post something similar regarding the impact of QE2 on economic data.