Charta is on Google Play: practise on the go. Get the app
Charta

Review what may be behind private credit growth and read regression results on how strongly each factor matters.

FRM® Part II · Current Issues in Financial Markets · The Global Drivers of Private Credit · FRM2.CI.3.b

The idea

Explanations for private credit growth split into supply and demand forces. On the supply side, tighter post-crisis bank capital and liquidity rules made riskier leveraged loans costlier for banks to hold. Low interest rates also pushed institutional investors toward higher-yielding, illiquid assets. On the demand side, mid-sized and sponsor-backed borrowers want speed, certainty and flexible terms. Studies test these ideas with panel regressions of private credit relative to GDP on country-level factors. Read each coefficient as the expected change in the dependent variable for a one-unit change in the factor, holding the others fixed. To test a factor, use the ratio of the coefficient to its standard error, which gives the t-statistic. t = coefficient ÷ standard error. At a 5 percent two-sided level the critical value is about 1.96. To compare factors, put them on the same scale with the next formula. Effect = coefficient × standard deviation. Significance shows association, not proof of causation, and a raw coefficient depends on the unit of the variable. Exam point: rank factors by coefficient times standard deviation, not by coefficient size.

Formulas

Economic effect of a one-standard-deviation move in a regression factor

effect = b × σ

b
the factor's estimated coefficient
σ
the standard deviation of the factor

Use it to rank factors measured in different units. Ranking by raw coefficient is the trap, and the sign gives direction only, not size.

t-statistic of a regression coefficient

t = b ÷ SE

b
the estimated coefficient
SE
its standard error

Compare with about 1.96 for a 5% two-sided test in a large sample. Significance shows association, not causation.

A practice question

From the question bank, with the reason each wrong option is wrong.

Which factor is most commonly proposed as a supply-side driver of the growth of private credit relative to bank lending?

  1. Looser bank capital requirements that made leveraged lending cheaper for banksWhy not: Wrong direction; looser rules would favour banks and slow the shift of lending to funds.
  2. A fall in institutional investor appetite for illiquid, higher-yielding assetsWhy not: Wrong direction; falling investor appetite would reduce, not raise, fund inflows.
  3. Tighter post-crisis bank capital and liquidity regulation that raised the cost to banks of holding riskier loans - correct
  4. A decline in mid-sized companies seeking financing outside public marketsWhy not: Describes lower borrower demand, which would slow growth, and it is a demand-side rather than supply-side factor.

Post-crisis regulation raised the capital cost of risky bank loans, leaving room for non-bank lenders to fill the gap. The other options either reverse the direction or concern demand.

A harder one

A panel regression of private credit assets (percent of GDP) on a bank capital tightness index (one unit on a 0 to 10 scale) with a large sample gives a coefficient of 0.42 and a standard error of 0.15. Which reading is correct?

  1. The t-statistic is 0.36, so the factor is not significantWhy not: Divides the standard error by the coefficient instead of the coefficient by the standard error.
  2. One more unit of tightness raises private credit assets by 42% of GDPWhy not: Misreads units; the coefficient is in percentage points of GDP, so 0.42 means 0.42 points, not 42.
  3. The t-statistic of 2.8 proves that tighter bank regulation caused the growth in private creditWhy not: Treats a significant association as proof of causation, ignoring omitted variables and reverse causality.
  4. The t-statistic is 2.8, significant at the 1% level; one more unit of tightness is associated with 0.42 percentage points higher assets to GDP - correct

t = 0.42 / 0.15 = 2.8, above the 5% critical value 1.96 and the 1% value 2.576. The coefficient is the associated change in percentage points for a one-unit change in the factor, holding others fixed.

The hardest tier

A regression of the change in private credit to GDP (percentage points) on changes in several factors gives a drift term of 0.5, a coefficient of 0.6 on the change in a bank capital tightness index, -0.3 on the change in the policy rate (percentage points) and 0.4 on the change in institutional assets to GDP (percentage points). The tightness index rises 1.5 units, the policy rate falls 2.0 points and institutional assets to GDP rise 1.0 point. The predicted change is:

  1. 1.9 percentage pointsWhy not: Omits the 0.5 intercept (0.9 + 0.6 + 0.4); the prediction includes the drift term.
  2. 1.2 percentage pointsWhy not: Applies the policy-rate change with the wrong sign (-0.3 x -2.0 taken as -0.6), giving 0.5 + 0.9 - 0.6 + 0.4.
  3. 2.4 percentage points - correct
  4. 2.0 percentage pointsWhy not: Drops the institutional assets term, giving 0.5 + 0.9 + 0.6 and ignoring a variable that moved.

Predicted change = 0.5 + 0.6 x 1.5 + (-0.3) x (-2.0) + 0.4 x 1.0 = 0.5 + 0.9 + 0.6 + 0.4 = 2.4. A negative coefficient times a fall in the factor adds to private credit.

In the app: 6 practice questions on this outcome, spaced review that brings it back before you forget it.

Start free →

More in The Global Drivers of Private Credit