## 1. Expected Pace of Policy Rates

## Source details

**Canonical URL:** [1. Expected Pace of Policy Rates](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1635.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1635.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1635.pdf.json)

---

### Background: Linked Exchange Rate System (LERS) and monetary mechanics
- Hong Kong dollar officially linked to the U.S. dollar at the rate of 7.8 Hong Kong dollars to one U.S. dollar since October 1983.
- System described as a Currency Board under which:
  - Interest rates adjust to inflows or outflows of funds.
  - The monetary base increases with capital inflow and contracts with capital outflow.
  - Expansion or contraction of the monetary base causes domestic interest rates to fall or rise, respectively, while the exchange rate remains stable.

### Recent interest-rate environment (historical context)
- For the past seven years, short-term interest rates in Hong Kong SAR have been near zero and long yields at near-historic lows, following the U.S. Fed lowering its federal funds rate to near zero with the onset of the global financial crisis.

### Recent policy shift and expectations
- Interest rates in Hong Kong SAR are expected to rise for the first time in about a decade as the U.S. Fed begins to increase the Federal Funds rate.
- On December 16, 2015, the Fed increased the target range for the Federal Funds rate by a quarter percentage point to 0.25‒0.50 percent, lifting it from the 0-0.25 percent range it has occupied since December 2008. Accordingly, the Hong Kong Monetary Authority (HKMA) also raised the base rate.

### Background and motivation
- Fed officials expected to increase the Federal Funds rate gradually over next years by carefully monitoring progress of actual inflation toward its inflation goal.
- Concern: whether the anticipated interest hike would be an additional large drag on growth in Hong Kong SAR given relatively subdued recent overall economic activity (GDP growth slowed to below 3 percent in 2014 and 2015).
- Key empirical questions posed:
  - How has private sector activity evolved during the previous rate hike cycles?
  - What are the main determinants for private consumption and corporate investment?
  - Would the impact be larger this time due to historic high level of leverage in both household and corporate sectors?

### Major findings (executive summary)
- Overall economic activity around the previous Fed tightening cycles remained robust on average, with only modest slowdown in domestic demand.
- Private sector leverage has increased to historic level in both households and corporate, but various strong buffers are in place that will likely contain the impact of higher interest rates.
- Higher mortgage payment burden is expected to have only limited impact on private consumption as nominal income of households would also rise under tight labor market conditions.
- Higher financing cost will also have only modest impact on corporate investment, unless accompanied by significant deterioration of demand outlook.

### I. Past episodes of Fed tightening (era of Hong Kong SAR’s currency board)
- Six rate-hike cycles since 1983 (when the currency board system was established).
- Over these six cycles, the Fed has increased the policy rate by about 2¾ percent over about 1¼ years on average.
- The pace of the latest rate hike starting in June 2004 was 4¼ percentage point increase over two years (noted as the slowest pace among cycles then).
- Fed "dot plot" (as of Dec. 2015) suggested the target policy rate would increase by about 1 percentage point a year over the next three years.
- Various interest rates in Hong Kong SAR have traced corresponding U.S. rates very closely across short-term money market rates to long-term lending rates.
- Historical pattern of private domestic demand around up-cycles:
  - Private consumption slowed somewhat right after rate hikes but recovered quickly from the second year.
  - Investment growth slowed a bit but continued to grow at robust pace after the rate hikes.

### II. Current economic setup in Hong Kong SAR — leverage and buffers
- Household sector:
  - Stock of household debt rose to historic highs, close to 70 percent of GDP, of which about two-thirds is residential mortgage debt.
  - HKMA’s estimate of the income-gearing ratio increased further to more than 70 percent, well above its long-term average of about 50 percent.
  - HKMA stress test: income-gearing ratio would rise to more than 90 percent if mortgage interest rates increase by 300 bps.
  - HKMA’s debt-service index of new mortgages would rise to more than 65 percent from the current level of less than 50 with a 300 bps rate rise.
  - New borrowers’ debt service relative to disposable income declined from 41 percent in August 2010 to 35 percent in 2014.
  - Number of owner-occupied households with mortgages or loans contracted by nearly one-fourth since the global financial crisis; only one-third of households currently carry mortgages.
  - Household debt servicing-to-income ratio would rise by about 5 percentage points with each 100 bps increase in interest rate (mechanical effect).
  - Historical mortgage payment-to-income responses: declined by 5-10 percent after previous Fed rate hikes in 1994 and 1999 respectively; increased by about 10 percentage points in 2004-05 when the policy rate rose by more than 4 percent.
- Corporate sector:
  - Total stock of bank lending to corporates increased to about 130 percent of GDP as of 2015Q2, from about 75 percent in early 2000s.
  - Corporate bond issuance expanded; new issuance of nonbank corporate debt securities increased by about seven-fold as of 2014 compared to 2007.
  - Cross-country firm-level indicators:
    - Leverage ratio of nonfinancial corporates in Hong Kong SAR is the lowest in the region (Figure 11).
    - Median short-term to long-term debt ratio is among the lowest in the region (Figure 12).
    - Less than 10 percent of total corporate debt is owed by firms with leverage ratio of more than 2 (Figure 13).
    - Less than 10 percent of total debt is owed by “stressed” companies with interest coverage ratio less than 1 (Figures 14–15).
  - Firm-level stress test: assuming average borrowing costs rise by 200 bps, just above 10 percent of total debt would be owed by “stressed” companies (interest coverage ratio less than 1).

### III. Empirical estimation — consumption
- Model specification:
  - Aggregate consumption estimated using an error-correction model (ECM) over 1994Q1 to 2015Q2:
    - Dependent variable: Cons = real consumption.
    - Explanatory variables: Income (proxy constructed by real payroll per worker × employment), Stock (real stock price), Prop (real property price), Mgt (mortgage payment-to-income ratio, Centraline measure).
    - All variables in logarithm except Mgt; augmented Dickey-Fuller tests indicate all series are I(1).
- Estimated short-run ECM result:
  - ΔCons_t = 0.016 ΔIncome_t − 0.075 ΔStock_t + 0.071 ΔProp_t + 0.090 ΔMgt_t + 0.000 ε_t − 0.106*(Cons_{t−1} − 0.355 Income_{t−1} + 0.200 Stock_{t−1} + 0.524 Prop_{t−1} + 0.521 Mgt_{t−1}) + ε_t
- Interpretation and numeric implications:
  - Consumption positively associated with income and wealth; negatively associated with debt servicing burden.
  - A 100 bps increase in mortgage rates mechanically raises mortgage payment-to-income ratio by about 5 percentage points (assuming no income growth).
  - Long-run elasticity of consumption with respect to debt service-to-income ratio estimated at about 0.5 → aggregate consumption would decline by about 2½ percent if other asset prices rose only at the rate of inflation.
  - Offsetting income effect: if nominal income grows about 5½ percent (in line with nominal GDP growth for 2016 in IMF’s WEO baseline) then households’ debt service-to-income ratio would rise by less than 2 percent against a 100 bps increase in mortgage rates.
  - With inflation about 3 percent a year, real income would rise by about 2½ percent a year.
  - Net implication: effect of higher interest rates through higher debt servicing burden would be largely offset by income effect; private consumption expected to continue rising at modest pace despite interest rate increases.

### IV. Empirical estimation — corporate investment
- Model specification:
  - Standard neoclassical investment model estimated with firm-level data (Worldscope) for about 1000 firms (1995–2014), excluding financial sector and firms with negative cash flows or negative leverage.
  - Investment I related to lagged investment, expectations of future profitability (Q), cash flow (CF), leverage (LEV), and lagged capital.
- Estimation results:
  - Fixed Effect Estimation:
    - I_{i,t}/K_{i,t−1} = 0.190 + 0.065 (I_{i,t−1}/K_{i,t−2}) + 0.032 Q_{i,t} + 0.043 (CF_{i,t}/K_{i,t−1}) + 0.051 LEV_{i,t} + ε_{i,t}
  - System GMM Estimation:
    - I_{i,t}/K_{i,t−1} = 0.244 + 0.069 (I_{i,t−1}/K_{i,t−2}) + 0.036 Q_{i,t} + 0.018 (CF_{i,t}/K_{i,t−1}) − 0.006 LEV_{i,t} + ε_{i,t}
- Interpretation and numeric implications:
  - Investment positively associated with Q and cash flow; lagged investment effect is positive and statistically significant.
  - Coefficient on leverage is not statistically significant in both methods (opposite signs across methods), implying leverage has not been a drag on investment in Hong Kong SAR.
  - Quantitative example: as of 2014, median effective interest rate in sample ≈ 4½ percent and median cash flow-to-capital stock ratio = 0.214.
    - A 100 bps increase in interest rates reduces median cash flow-to-capital stock ratio to 0.200.
    - With coefficient estimates on cash flow-to-capital stock ratio of 0.018 (system-GMM) to 0.043 (fixed effect) → corporate investment would decline by 1/8 to 1/3 percent, assuming no change in other factors (expected profitability, leverage).
  - Actual financing conditions may not tighten as much if major central banks maintain expanded balance sheets.

### V. Tail-risk scenario — “perfect storm”
- Risks that could amplify impact:
  - More aggressive portfolio rebalancing away from emerging market assets to U.S. dollar assets → rapid decompression of spreads globally.
  - Faster-than-expected U.S. inflation/wage-price dynamics → Fed Funds rate closer to upper bound of FOMC forecasts → upward shift of entire yield curve in Hong Kong SAR.
  - China growth slowing more sharply than expected → asset markets adjust sharply.
- Historical analogs:
  - Asian financial crisis (late 1990s) and global financial crisis saw stock and property prices decline significantly (example: about 30 percent declines in 1998 for both asset prices).
  - Payroll growth in real terms was broadly flat during those episodes.
- Quantified tail-risk impact:
  - If stock and house prices decline by 20 to 30 percent and interest rates rise by 200 basis points simultaneously, the contribution of domestic demand to growth would fall by 3 to 5 percentage points compared to the baseline.
  - Prior study: hard landing in Mainland China could reduce Hong Kong SAR’s output by as much as about 3 percentage points below baseline in the first two years, pushing Hong Kong SAR into recession and possible deflation.

### Conclusion and policy implications
- With anticipated gradual increases in interest rates and buffers in place, private domestic activity (both consumption and investment) is expected to remain robust.
- Key mechanisms:
  - Income growth (tight labor market) is projected to offset part of the higher debt service burden on households.
  - Corporate leverage quality and favorable maturity structure suggest limited sensitivity of investment to modest interest rate increases.
- Caveat: a significant global shock (rapid rise in rates, sharp China slowdown, large asset price corrections) could materially weaken private domestic demand and investment.

*Source: _wp1635 - 1. Expected Pace of Policy Rates (excerpt) — IMF PDF*

### 1. Expected Pace of Policy Rates  ......................................................................................

### Expected Pace of Policy Rates

### Background: Linked Exchange Rate System (LERS) and monetary mechanics
- Hong Kong dollar officially linked to the U.S. dollar at the rate of 7.8 Hong Kong dollars to one U.S. dollar since October 1983.
- System described as a Currency Board under which:
  - Interest rates adjust to inflows or outflows of funds.
  - The monetary base increases with capital inflow and contracts with capital outflow.
  - Expansion or contraction of the monetary base causes domestic interest rates to fall or rise, respectively, while the exchange rate remains stable.

### Recent interest-rate environment (historical context)
- For the past seven years, short-term interest rates in Hong Kong SAR have been near zero and long yields at near-historic lows, following the U.S. Fed lowering its federal funds rate to near zero with the onset of the global financial crisis.

### Recent policy shift and expectations
- Interest rates in Hong Kong SAR are expected to rise for the first time in about a decade as the U.S. Fed begins to increase the Federal Funds rate.
- On December 16, 2015, the Fed increased the target range for the Federal Funds rate by a quarter percentage point to 0.25‒0.50 percent, lifting it from the 0-0.25 percent range it has occupied since December

*Source: _wp1635 - 1. Expected Pace of Policy Rates (excerpt) — IMF PDF*

### 2008. Accordingly, the Hong Kong Monetary Authority (HKMA) also raised the base rate—

### _wp1635 - 2008. Accordingly, the Hong Kong Monetary Authority (HKMA) also raised the base rate—

### Background and motivation
- Fed officials expected to increase the Federal Funds rate gradually over next years by carefully monitoring progress of actual inflation toward its inflation goal.
- Concern: whether the anticipated interest hike would be an additional large drag on growth in Hong Kong SAR given relatively subdued recent overall economic activity (GDP growth slowed to below 3 percent in 2014 and 2015).
- Key empirical questions posed:
  - How has private sector activity evolved during the previous rate hike cycles?
  - What are the main determinants for private consumption and corporate investment?
  - Would the impact be larger this time due to historic high level of leverage in both household and corporate sectors?

### Major findings (executive summary)
- Overall economic activity around the previous Fed tightening cycles remained robust on average, with only modest slowdown in domestic demand.
- Private sector leverage has increased to historic level in both households and corporate, but various strong buffers are in place that will likely contain the impact of higher interest rates.
- Higher mortgage payment burden is expected to have only limited impact on private consumption as nominal income of households would also rise under tight labor market conditions.
- Higher financing cost will also have only modest impact on corporate investment, unless accompanied by significant deterioration of demand outlook.

### I. Past episodes of Fed tightening (era of Hong Kong SAR’s currency board)
- Six rate-hike cycles since 1983 (when the currency board system was established).
- Over these six cycles, the Fed has increased the policy rate by about 2¾ percent over about 1¼ years on average.
- The pace of the latest rate hike starting in June 2004 was 4¼ percentage point increase over two years (noted as the slowest pace among cycles then).
- Fed "dot plot" (as of Dec. 2015) suggested the target policy rate would increase by about 1 percentage point a year over the next three years.
- Various interest rates in Hong Kong SAR have traced corresponding U.S. rates very closely across short-term money market rates to long-term lending rates.
- Historical pattern of private domestic demand around up-cycles:
  - Private consumption slowed somewhat right after rate hikes but recovered quickly from the second year.
  - Investment growth slowed a bit but continued to grow at robust pace after the rate hikes.

### II. Current economic setup in Hong Kong SAR — leverage and buffers
- Household sector:
  - Stock of household debt rose to historic highs, close to 70 percent of GDP, of which about two-thirds is residential mortgage debt.
  - HKMA’s estimate of the income-gearing ratio increased further to more than 70 percent, well above its long-term average of about 50 percent.
  - HKMA stress test: income-gearing ratio would rise to more than 90 percent if mortgage interest rates increase by 300 bps.
  - HKMA’s debt-service index of new mortgages would rise to more than 65 percent from the current level of less than 50 with a 300 bps rate rise.
  - New borrowers’ debt service relative to disposable income declined from 41 percent in August 2010 to 35 percent in 2014.
  - Number of owner-occupied households with mortgages or loans contracted by nearly one-fourth since the global financial crisis; only one-third of households currently carry mortgages.
  - Household debt servicing-to-income ratio would rise by about 5 percentage points with each 100 bps increase in interest rate (mechanical effect).
  - Historical mortgage payment-to-income responses: declined by 5-10 percent after previous Fed rate hikes in 1994 and 1999 respectively; increased by about 10 percentage points in 2004-05 when the policy rate rose by more than 4 percent.
- Corporate sector:
  - Total stock of bank lending to corporates increased to about 130 percent of GDP as of 2015Q2, from about 75 percent in early 2000s.
  - Corporate bond issuance expanded; new issuance of nonbank corporate debt securities increased by about seven-fold as of 2014 compared to 2007.
  - Cross-country firm-level indicators:
    - Leverage ratio of nonfinancial corporates in Hong Kong SAR is the lowest in the region (Figure 11).
    - Median short-term to long-term debt ratio is among the lowest in the region (Figure 12).
    - Less than 10 percent of total corporate debt is owed by firms with leverage ratio of more than 2 (Figure 13).
    - Less than 10 percent of total debt is owed by “stressed” companies with interest coverage ratio less than 1 (Figures 14–15).
  - Firm-level stress test: assuming average borrowing costs rise by 200 bps, just above 10 percent of total debt would be owed by “stressed” companies (interest coverage ratio less than 1).

### III. Empirical estimation — consumption
- Model specification:
  - Aggregate consumption estimated using an error-correction model (ECM) over 1994Q1 to 2015Q2:
    - Dependent variable: Cons = real consumption.
    - Explanatory variables: Income (proxy constructed by real payroll per worker × employment), Stock (real stock price), Prop (real property price), Mgt (mortgage payment-to-income ratio, Centraline measure).
    - All variables in logarithm except Mgt; augmented Dickey-Fuller tests indicate all series are I(1).
- Estimated short-run ECM result (presented in the source):
  - ΔCons_t = 0.016 ΔIncome_t − 0.075 ΔStock_t + 0.071 ΔProp_t + 0.090 ΔMgt_t + 0.000 ε_t − 0.106*(Cons_{t−1} − 0.355 Income_{t−1} + 0.200 Stock_{t−1} + 0.524 Prop_{t−1} + 0.521 Mgt_{t−1}) + ε_t
- Interpretation and numeric implications:
  - Consumption positively associated with income and wealth; negatively associated with debt servicing burden.
  - A 100 bps increase in mortgage rates mechanically raises mortgage payment-to-income ratio by about 5 percentage points (assuming no income growth).
  - Long-run elasticity of consumption with respect to debt service-to-income ratio estimated at about 0.5 → aggregate consumption would decline by about 2½ percent if other asset prices rose only at the rate of inflation.
  - Offsetting income effect: if nominal income grows about 5½ percent (in line with nominal GDP growth for 2016 in IMF’s WEO baseline) then households’ debt service-to-income ratio would rise by less than 2 percent against a 100 bps increase in mortgage rates.
  - With inflation about 3 percent a year, real income would rise by about 2½ percent a year.
  - Net implication: effect of higher interest rates through higher debt servicing burden would be largely offset by income effect; private consumption expected to continue rising at modest pace despite interest rate increases.

### IV. Empirical estimation — corporate investment
- Model specification:
  - Standard neoclassical investment model estimated with firm-level data (Worldscope) for about 1000 firms (1995–2014), excluding financial sector and firms with negative cash flows or negative leverage.
  - Investment I related to lagged investment, expectations of future profitability (Q), cash flow (CF), leverage (LEV), and lagged capital.
- Estimation results (two methods reported):
  - Fixed Effect Estimation:
    - I_{i,t}/K_{i,t−1} = 0.190 + 0.065 (I_{i,t−1}/K_{i,t−2}) + 0.032 Q_{i,t} + 0.043 (CF_{i,t}/K_{i,t−1}) + 0.051 LEV_{i,t} + ε_{i,t}
  - System GMM Estimation:
    - I_{i,t}/K_{i,t−1} = 0.244 + 0.069 (I_{i,t−1}/K_{i,t−2}) + 0.036 Q_{i,t} + 0.018 (CF_{i,t}/K_{i,t−1}) − 0.006 LEV_{i,t} + ε_{i,t}
- Interpretation and numeric implications:
  - Investment positively associated with Q and cash flow; lagged investment effect is positive and statistically significant.
  - Coefficient on leverage is not statistically significant in both methods (opposite signs across methods), implying leverage has not been a drag on investment in Hong Kong SAR.
  - Quantitative example: as of 2014, median effective interest rate in sample ≈ 4½ percent and median cash flow-to-capital stock ratio = 0.214.
    - A 100 bps increase in interest rates reduces median cash flow-to-capital stock ratio to 0.200.
    - With coefficient estimates on cash flow-to-capital stock ratio of 0.018 (system-GMM) to 0.043 (fixed effect) → corporate investment would decline by 1/8 to 1/3 percent, assuming no change in other factors (expected profitability, leverage).
  - Actual financing conditions may not tighten as much if major central banks maintain expanded balance sheets.

### V. Tail-risk scenario — “perfect storm”
- Risks that could amplify impact:
  - More aggressive portfolio rebalancing away from emerging market assets to U.S. dollar assets → rapid decompression of spreads globally.
  - Faster-than-expected U.S. inflation/wage-price dynamics → Fed Funds rate closer to upper bound of FOMC forecasts → upward shift of entire yield curve in Hong Kong SAR.
  - China growth slowing more sharply than expected → asset markets adjust sharply.
- Historical analogs:
  - Asian financial crisis (late 1990s) and global financial crisis saw stock and property prices decline significantly (example: about 30 percent declines in 1998 for both asset prices).
  - Payroll growth in real terms was broadly flat during those episodes.
- Quantified tail-risk impact:
  - If stock and house prices decline by 20 to 30 percent and interest rates rise by 200 basis points simultaneously, the contribution of domestic demand to growth would fall by 3 to 5 percentage points compared to the baseline.
  - Prior study: hard landing in Mainland China could reduce Hong Kong SAR’s output by as much as about 3 percentage points below baseline in the first two years, pushing Hong Kong SAR into recession and possible deflation (N’Diaye and Ahuja, 2012 referenced).

### Conclusion and policy implications
- With anticipated gradual increases in interest rates and buffers in place, private domestic activity (both consumption and investment) is expected to remain robust.
- Key mechanisms:
  - Income growth (tight labor market) is projected to offset part of the higher debt service burden on households.
  - Corporate leverage quality and favorable maturity structure suggest limited sensitivity of investment to modest interest rate increases.
- Caveat: a significant global shock (rapid rise in rates, sharp China slowdown, large asset price corrections) could materially weaken private domestic demand and investment.

*Source: IMF working paper content unit _wp1635 - 2008. Accordingly, the Hong Kong Monetary Authority (HKMA) also raised the base rate—*

### REFERENCES

### _wp1635 - REFERENCES

### Academic articles and books
- Bloom, N., S. Bond, and J. Van Reenen, 2007, “Uncertainty and Investment Dynamics” Review of Economic Studies, Vol. 74, pp. 391–415.
- Christiano, L., Eichenbaum, M., Evans, C., 2005, “Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy,” Journal of Political Economy, Vol. 113, pp. 1‒45.
- Eberly J., S. Rebelo, and N. Vincent, 2012, “What Explains the Lagged-Investment Effect?” Journal of Monetary Economics, Vol. 59, pp. 370–380.
- Fazzari, S., G. Hubbard, and B. Petersen, 1988, “Financing Constraints and Corporate Investment,” Brookings Papers on Economic Activity, pp. 141–195.
- Hayashi, F., 1982, “Tobin’s Marginal Q and Average Q: A Neoclassical Interpretation,” Econometrica, Vol. 50, pp. 215–224.
- Hubbard, R.G., 1998, “Capital-Market Imperfections and Investment,” Journal of Economic Literature, Vol. 36, pp. 193–225.
- Kang, W., K. Lee, and R. A. Ratti, 2014, “Economic Policy Uncertainty and Firm-Level Investment,” Journal of Macroeconomics, Vol. 39, pp. 42–53.

### Institutional reports and working papers
- Hong Kong Monetary Authority, 2015, Half-Yearly Monetary and Financial Stability Report (Hong Kong: September)
- International Monetary Fund, 2014, Regional Economic Outlook: Asia and the Pacific (Washington, April).
- Kang, J. S. and S. Piao, “Production Offshoring and Investment by Japanese Firms,” IMF Working Paper, No. 15/183 (Washington: International Monetary Fund).
- N'Diaye, P. and A. Ahuja, “Trade and Financial Spillover on Hong Kong SAR from a Downturn in Europe and Mainland China,” IMF Working Paper, No. 12/81 (Washington: International Monetary Fund).

*Source: _wp1635 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1635.pdf_