Credit impulse background


13 Jun 2011.

Spending is a flow, and should be compared with net new lending, a flow, rather than credit outstanding, a stock. Hence changes in spending are dependent on changes in net new lending or, equivalently, on the acceleration of credit.

Michael Biggs, economist at Deutsche Bank, introduced this idea in Nov 2008 and, by analogy with the well known concept of fiscal impulse, defined “credit impulse” as the “the change in new credit issued as a % of GDP”. Since 2008, it has been shown, for many time periods and many countries, that private sector demand is very closely correlated with private sector credit impulse. This explains a number of conundrums missed by those who focus on the credit stock.


20 May 2013. Data through Q1 2012.


Calculation (14 Jun 2011) The credit impulse is considered for the private sector (with any government fiscal impulse analyzed separately). It is calculated as follows:

  1. Take quarterly growth in nominal credit of the non-financial private sector
  2. Normalize by nominal GDP for the quarter over which the credit growth is measured
  3. Take the year-over-year actual (not percent) change in that normalized quantity
  4. Multiply by 100

This is then compared to the year-over-year percent change in consumption (PCE) plus investment (GPDI).

Money supply and GDP (14 Jun 2011) By arguments similar to those supporting the credit impulse, Biggs also argues that nominal GDP growth is more closely related to changes in M3 flows than to changes in M3. An empirical check on European data supports the conclusion.

Purchase can directly create credit (17 Jun 2011) It is obvious that borrowing can enable lending. Perhaps slightly less obvious that a purchase, if made on credit, can simultaneously create debt on the purchaser's account and credit on the seller's account.

Simple illustration (13 Jun 2011) Consider an economy in which the household sector is one person.

  • If someone is borrowing a fixed amount each month then spending is constant. So there is growth in credit but no growth in impulse or in spending.
  • If that person borrows more each month, then credit is accelerating, and impulse and spending growth are positive.
  • If a person is paying down debt with a constant payment each month, credit is falling, but impulse and spending growth are zero.
  • If that person continues to pay down debt each month, but by decreasing amounts, then credit is falling, but impulse and spending growth are positive.

See also

Selected commentary

Clippings below were used in the construction of this page

Covered through 11 Jun 2011.

Credit impulse and growth in spending

19 Nov 2008. Global Macro Issues newsletter from Deutsche Bank.

“The impact of credit on growth. Michael Biggs”

“The conventional measure used when associating developments in credit with developments in domestic demand is credit growth. This is growth in the stock of credit. In our view, domestic demand depends on the amount borrowed in a particular period, or the flow of credit. If this is correct2, then growth in domestic demand is a function of growth in new credit issued, and not growth in credit. The impact of credit on demand will also depend on the amount of credit extended relative to the size of the economy, and consequently our preferred credit measure is the change in new credit issued as a % of GDP. For the remainder of this piece we refer to this measure as the “credit impulse”.

For the sake of clarity, assume

ct = Ct – Ct-1

where Ct is the stock of credit at the end of period t, and ct is the net new borrowing or the net amount of new credit issued during period t. Our estimate of the credit impulse CIt is therefore:

CIt = ct/GDPt – ct-1/GDPt-1

This measure is the private sector equivalent to the fiscal impulse, which is, often measured as the change in the structural budget deficit as a % of GDP. The fiscal deficit is the flow of credit going to the public sector, while ct is the flow of credit going to the private sector. And, for the same reasons that the impact of fiscal policy on domestic demand is measured by looking at the fiscal impulse rather than growth in the public sector debt, the appropriate private credit measure is the credit impulse rather than credit growth.

To differentiate between the credit impulse and the fiscal impulse, we use only private credit for our credit impulse measure. This is obtained from the flow of funds by taking total credit to the domestic non-financial sectors and subtracting credit to the state, local and federal governments. …

[For the US] Both series [credit growth and credit impulse] are adequately correlated with domestic demand from 1974 – 1990. In 1992 and 1993, however, the sharp rebound in domestic demand growth is far more consistent with the spike in the credit impulse than the moderate recovery in credit growth. Real credit growth has remained strong since the late 1990s. The credit impulse, in contrast, fell sharply in 2001, increased to a peak in 2004, and has declined ever since. This is exactly what has happened to domestic demand growth. Not only is the credit impulse measure intuitively superior to credit growth, but it is also better correlated with developments in domestic demand. …

A number of countries experienced a sharp slowdown in GDP growth in the early 1990s. Some (US, UK, Sweden) experienced a more rapid recovery, while others (Germany, Japan) experienced protracted periods of subdued demand and weak growth. In our view, one of the key differentiating factors between these two groups was the speed of the fall in credit growth as economic activity started to slow. The former experienced a sharp initial fall in credit growth, followed by a mild rebound that was enough to cause a recovery in the credit impulse. The latter, in contrast, experienced only a very slowdown in credit growth. The conditions were never created in which the credit impulse could rebound, and consequently domestic demand remained weak.”

Credit impulse and demand in New Zealand

24 Nov 2008. ANZ Bank newsletter.

“Market focus New Zealand”

“In regards to credit growth, we were forwarded an interesting piece from an offshore investment bank noting the material differences between the rate of credit growth, and the change in the rate of credit growth (Acknowledgement needs to go to Michael Biggs from Deutsche Bank for the framework and analysis on the US.). People tend to focus on the former, although the relationship with final demand is relatively loose. A purer measure, which the researchers defined at the “credit impulse”, is simply the change in the stock of credit to GDP compared to last period.

CI(t) = c(t)/GDP(t) - c(t-1)/GDP(t-1)

where CI(t) is the credit impulse and c(t) is the change in the stock of credit. Using this approach we can map a pretty close relationship to final domestic demand.”

Application to 1930s

13 Mar 2009. Deutsche Bank Global Macro Issues (private communication).

“Credit impulse: Lessons from the Great Depression. Michael Biggs and Thomas Mayer”

“The data from the US in the 1930s support our view - credit stock growth in the US was negative from 1931 - 1935, but the credit impulse was strongly positive in 1934 and 1935. Private sector real domestic demand growth reached +9.3% in 1934 and +10.1% in 1935, even though credit stock growth was negative.”

Details of the calculation

Jul 2009. De Nederlandsche Bank Working Paper 218.

“Credit and economic recovery. Michael Biggs, Thomas Mayer and Andreas Pick”

“Ideally, the credit measures should be based on the broadest possible measure of credit, and should capture only credit extended to the non- financial private sector. Such data are easily available in the countries that produce flow of funds statements (US, Japan, and the UK) but not in most others. For those without flow of funds statements, we use net credit extended by the banking sector to non-financial corporations, households, and non-profit institutions. Furthermore, we measure only credit extended to the private sector and, where possible, plot this against private sector real domestic demand growth. …

For the analysis of the post-war U.S. data we use quarterly data on real consumption, real investment, nominal GDP, nominal credit and the GDP deflator … Private demand growth as

[YOY percent change in real consumption + real investment]

… Dt is nominal credit and Pt is the GDP deflator. Finally, the credit impulse is defined as

cti = 100 [ (Dt-D(t-1))/Ytn - (D(t-4)-D(t-5)/Y(t-5)n ]

where Ytn is nominal GDP.”

[I think that Y(t-5)n should be Y(t-4)n. If so, then the credit impulse is formed as follows:

  1. Take quarterly growth in credit of the non-financial private sector
  2. Normalize by nominal GDP for the quarter over which the credit growth is measured
  3. Take the year-over-year actual (not percent) change in that normalized quantity
  4. Multiply by 100

This is then compared to the year-over-year percent change in consumption (PCE) plus investment (GPDI).]

The myth of the "Phoenix Miracle"

14 May 2010. VOX EU.

“The myth of the “Phoenix Miracle”. Michael Biggs, Thomas Mayer, Andreas Pick”

“The observation that economies can recover from a crisis without the need for credit growth is known as a “Phoenix Miracle”. This column argues that this theory is based on an inappropriate comparison between GDP – a flow variable – and the stock of credit. If GDP is instead compared with the flow of credit, it is evident that GDP and credit recover simultaneously.”

Nominal GDP follows M3 impulse, not M3 growth

7 Jul 2010. Deutsche Bank Global Economic Perspectives

On Scribd

“MV=PQ, and why we are not concerned about the contraction in money. Michael Biggs”

“In the previous section, we argued: 1) that monitoring developments in money was equivalent to monitoring developments in credit, and 2) that the important credit variable to follow was the change in the flow of credit rather than the growth in the stock. Taken together, this implies that what we should be following is changes in the flow of money rather than growth in the money stock.

This view finds tentative support in the data for the euro area. For the period 1995 – 2010 (the entire period for which we have data), the correlation between nominal GDP growth and M3 growth is 0.42, whereas the correlation between nominal GDP growth and changes in the flow of M3 is 0.76 (0.79 for the flow of credit). …

Credit versus money

We argued in this piece that monitoring M3 was equivalent to monitoring credit. To make this point more strongly –– we would prefer to monitor private sector credit than M3, because we believe that developments in credit give one a more accurate picture of developments in the underlying economy. If, for example, households become more risk averse and want quicker access to their money, they may choose to put their deposits on immediate notice rather than three months notice. If this were the case, M3 would rise, longer term financial liabilities would fall, and credit would be unchanged. The surge in M3 would not be reflected in developments in the underlying economy – these would probably be running more in line with the unchanged credit.”

[Given this, one should re-evaluate whether inflation is related more to changes in M2 or M2 impulse.]

Purchases create credit "out of thin air"

11 Jun 2011. Steve Keen Debtwatch

“Dude, where's my recovery?”

“as long as 4 decades ago, the actual situation was put very simply by the then Senior Vice President, Federal Reserve Bank of New York, Alan Holmes. Holmes explained why the then faddish Monetarist policy of controlling inflation by controlling the growth of Base Money had failed, saying that it suffered from “a naive assumption” that:

the banking system only expands loans after the [Federal Reserve] System (or market factors) have put reserves in the banking system. In the real world, banks extend credit, creating deposits in the process, and look for the reserves later. The question then becomes one of whether and how the Federal Reserve will accommodate the demand for reserves. In the very short run, the Federal Reserve has little or no choice about accommodating that demand; over time, its influence can obviously be felt. (Alan R. Holmes, 1969, p. 73)

The empirical fact that “loans create deposits” means that the change in the level of private debt is matched by a change in the level of money, which boosts aggregate demand.”

And going a step further:

“all demand is monetary, and there are two sources of money: incomes, and the change in debt. … Aggregate demand is therefore equal to Aggregate Supply plus the change in debt.

Thirdly, this Aggregate Demand is expended not merely on new goods and services, but also on net sales of existing assets. …

Income + Change in Debt = Output + Net Asset Sales”