Econ 0609

HAWAII ECONOMIC TRENDS June 30, 2009

prepared by

Paul H. Brewbaker, Senior Economic Advisor Bank of Hawaii

https://www.boh.com/econ/512_539.asp

Economic Summary June 30, 2009 •

From the economic freefall that was underway in Hawaii during fourth quarter 2008 a transition to emerging stabilization began to appear in second quarter 2009. As a reminder that progress towards economic recovery will be uneven, the biological surprise of the H1N1-A influenza pandemic late in April 2009 added new economic uncertainty to Hawaii’s economic outlook. Historically, biological events such as the SARS episode in 2003, which set back international travel to Hawaii almost as much as the 9/11 event, have had the potential to undermine the economic forecast. Coming just as this report was to be published in early-May, the H1N1 pandemic required a tourism forecast recalibration. At least one published Hawaii economic forecast had to be revised in the wake of H1N1; this report takes into account about two months of daily international passenger arrivals data from late-April through late-June to take a first stab at a tourism forecast revision.

•

The dominance of tourism in Hawaii’s changing fortunes during the last four quarters was not just the consequence of tourism’s importance as the island economy’s principal export, comprising 15-20 percent of Hawaii value-added. Tourist volumes were moving more than anything else in Hawaii’s economy (except petroleum prices) for the last five quarters. Unlike the early-1990s, when domestic air lift to Hawaii was gradually reduced by roughly one-quarter, the shutdown of Aloha Airlines and ATA in March/April 2008 instantaneously reduced capacity by nearly one-fifth. Rising oil prices and consumer retrenchment meant that this lift was not replaced, and the April 2008 tourism forecast—which anticipated a “9/11-like” impact—proved to be premature: the sharp initial drop in April’s forecast for tourist volume more than doubled by the time of the November 2008 forecast revision.

•

The sharper “9/11-type” decline in 2008 travel to Hawaii means that the tourism recovery extends to several more years than previously envisioned in the April 2008 forecast. Airline shutdowns, consumer recession, and the post-Lehman Brothers financial panic all contributed. In the end, 2008 Hawaii visitor arrivals (by air) declined 12.2 percent on the domestic side, and 6.1 percent on the international side. In first quarter 2009 domestic and international arrivals were down 17.7 percent and 5.2 percent, respectively.

•

Second quarter 2009 Hawaii tourism performance entered the window “one-year-since” the Aloha/ATA shutdowns occurred in spring 2008. From this point forward, comparisons to one year earlier will exhibit more muted losses; domestic arrivals changed −3.6 percent in April and −2.2 percent in May 2009, yearover-year, and turned upward in summer. Official data are not seasonally-adjusted but, in fact, the trend for seasonally-adjusted domestic arrivals since late-summer 2008 actually has been flat to up slightly. This rising pattern continued in spring 2009. Hawaii daily international passenger arrivals growth, April 2009 through June 26, 2009 (percent change, year-over-year, from comparable days of the week one year earlier; the H1N1-A influenza virus gradually emerged, eventually to global pandemic status, beginning in the final weeks of April 2009)

40% 20% 0% -20% -40% 4/1/09 2

5/2/09

6/2/09

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

Bank of Hawaii semi-annual Business Confidence Survey results: diffusion index

University of Michigan Index of U.S. consumer sentiment

(positive sentiment above zero; neutral equals zero)

(monthly, first quarter 1966 = 100)

120

100

110

50

100 90

0 80 70

-50

60

-100

NBER recessions shaded

1990

1995

2000

2005

2010

50 NBER recessions shaded 90 95 00

05

10

•

International travel experienced a sudden but temporary drop after the collapse of Lehman Brothers in September 2008, but by year-end was rebounding on a seasonally-adjusted basis. Stabilization was beginning to appear in seasonally-adjusted data until the H1N1 virus appeared. Now, after eight to ten weeks of daily international arrivals data (graph at bottom of page 2) it seems as if the H1N1 pandemic is not having as dramatic an impact as did SARS six years ago. Thankfully, the widespread incidence of this viral pandemic does not appear to have been matched by its severity. The new tourism forecast in this report anticipates shrinking losses in international travel (reported on page 4).

•

Just as the transition through economic stabilization in mid-2009 to economic recovery at year end is forecast for annualized U.S. real GDP growth, the transition from negative to positive year-over-year growth began in domestic passenger arrivals in May and most of June 2009. The long, slow tourism recovery will be damped by challenges on the international side. Consensus Hawaii tourism forecasts anticipate improvement from −12 percent domestic arrivals growth through May 2009. Second half gains should be sufficient for domestic annual arrivals growth of −3 percent; gains will extend into 2010. Views are less uniform regarding the international arrivals forecast, but modest cumulative losses of 6 to 9 percent in international arrivals from now trough end-2010 are expected.

•

The 2008-2009 recession continued to play out in Hawaii economic data through mid-year. Hawaii unemployment through the end of second quarter 2009 hovered at a seasonally-adjusted 7.4 percent of the labor force, about two percentage points below the national average, and about a percentage point below the forecasts for Hawaii in 2010 included in this report. Private job loss this cycle was particular sharp, worse than in past downturns, although the annualized growth rate of seasonally-adjusted payroll employment did decelerate from −3.35 percent in December 2008 to −2.91 percent through May 2009. Public sector workforce reduction is expected to extend economy-wide job loss through year-end 2009 in Hawaii, which will likely see continued employment erosion even after the recovery has begun.

•

Sentiment probably began to shift from negative to neutral in second quarter 2009, based on surveys and the revival of investor interest in global equity markets earlier in the spring. Economic forecasts between second quarter 2009 and first quarter 2009 were not revised very much, also marking a widening perception of stabilization when compared to downward forecast revisions late in 2008. Recovery can only come after the economic declines have ended—though lagging indicators (employment, foreclosures, bankruptcies) will remain fodder for pessimism. Just as Hawaii went into the downturn in step with the U.S. mainland, so is it likely that economic recovery by end-2009 will push through in 2010 and beyond, both in the islands as well as nationwide.

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

3

Comparative Hawaii economy 2009-2010 forecasts June 30, 2009

Jobs

Personal Income CPI (real) Honolulu 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010

Annual percent changes 1

Brewbaker 2 UHERO DBEDT Laney

3

Actual(p)

-0.4

-2.9

-1.3

0.9

1.8

-0.3

-2.5

-1.1

4.6

4.6

4.6

-0.2

1.2

2.6

2.1

0.0

-2.9

-0.6

1.1

1.6

-0.2

-2.7

-0.6

5.1

5.0

4.5

0.5

0.3

1.9

0.0

-2.1

0.0

1.8

-0.2

-1.1

0.0

4.5

4.2

1.2

1.5

2.5

2.0

-0.3

-1.2

2.0

1.0

-1.0

-1.5

5.0

5.0

5.0

3.5

2.6

1.3

-0.9

1.4

1.0

0.2

5.9

4.8

4.3

Total Domestic International Visitor (UHERO: US) (UHERO: Japan) Arrivals by air Arrivals by air Arrivals by air 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010

Annual percent changes 1

Brewbaker UHERO

Laney

1.9

4

6

DBEDT

2.5

2

3

-0.5

-1.6 -10.6

-3.1

3.7

1.7

-0.1 -12.6

-2.7

5.7

-5.9

-5.3

-4.7

-4.4

-1.8

0.4

-1.1 -10.8

-6.8

3.1

2.4

-0.8 -14.2

-3.2

1.7

-8.6

-3.4

-9.8 -13.8

5.7

-0.8 -10.1

-5.9

1.2

0.5

-0.5

-5.0

0.6

-1.2 -10.6

3.3

0.0 -12.2

-4.8

-4.7

-6.1

4

6

Actual(p)

-9.0

Total Annual percent changes 1

Visitor Unemployment rate Construction5 Expenditures (UHERO ; see footnote) (UHERO ; see footnote) 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010

Brewbaker 2 UHERO

3

DBEDT 4 Laney

2.0

-6.4

6

Actual(p)

4.0

2.6 -11.4

-7.9

4.0

7.9

8.4 8.1

2.7

2.5

3.9

7.4

2.8

2.5

4.2

5.5

2.5

2.7

8.9

-1.2

0.2 -11.2 -15.5 -3.7 -19.4 -10.3

5.2

12.6

11.7

5.8

Notes: 1

2006-2008 "forecasts" are taken from last published forecasts prior to end of each year

2

Paul Brewbaker (TZ Economics), Senior Economic Advisor, Bank of Hawaii (https://www.boh.com/econ/512_803.asp); construction forecasts are unpublished, as prepared for Hawaii Council on Revenues

3

Professors Carl Bonham and Byron Gangnes (University of Hawaii Economic Research Organization), "UHERO Hawai'i Quarterly Forecast Update: State Budget Crisis Threatens Recovery," (June 12, 2009) (http://www.uhero.hawaii.edu/eis/eis_forecastarchive.html) 4

Hawaii DBEDT, "Outlook for the Economy: 2nd Quarter 2009," (May 18, 2009) (DBEDT visitor estimates include cruise ship arrivals) http://hawaii.gov/dbedt/info/economic/data_reports/info/economic/data_reports/qser/outlook-economy 5

Carl Bonham (UHERO) and Paul Brewbaker, "UHERO Hawai'i Construction Forecast: Global Downturn Hammers Construction" (March 6, 2009) (http://www.uhero.hawaii.edu/eis/eis_forecastarchive.html)

6

Professor Leroy Laney, Hawaii Pacific University (https://www.fhb.com/hm_econ.htm)

Data were compiled by TZ Economics but users are encouraged to refer to orginal sources for updated information.

4

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

5

Hawaii Economic Indicators

644.62 628.93 2.5 626.96 624.69 619.96 976.3 443.9 165.8 363.6 156.6 2,431 984 1,447 627.7 317.2 1,887.3 1,409.9 477.4 79.3 197.51

thousands thousands percent thousands thousands thousands million 2008$ million 2008$ million 2008$ million 2008$ million 2008$ units units units thousand $ thousand $ thousand thousand thousand percent 2007$

Civilian labor force Civilians employed Civilian unemployment rate Total wage and salary jobs pre-benchmarked jobs Nonagricultural jobs

Private building permits Residential Commercial and industrial Additions and alterations Government contracts

Oahu home sales Single family: Honolulu Condominium: Honolulu SF median price: Honolulu Condo median price: HNL

Visitor arrivals Domestic arrivals International arrivals Hotel occupancy Average daily room rate

*interpolation; 2009:1 forecast

2006:3 51.992 2.0 5.84

billion 2008$ % y-o-y % y-o-y

Real personal income Growth rate Honolulu CPI inflation*

Seasonally-adjusted, real data (home prices nominal)

2,500 1,000 1,500 633.4 321.9

814.5 435.6 97.6 245.5 157.7

648.04 632.08 2.4 630.06 632.18 623.31

2007:1 52.345 1.2 5.24

2,453 979 1,474 650.0 325.0

930.2 557.6 195.9 180.3 327.5

645.13 629.52 2.5 630.70 634.31 624.12

2007:2 52.547 0.8 4.95

2,157 881 1,276 642.9 327.6

819.7 398.1 113.1 318.6 184.9

644.18 626.57 2.7 631.98 636.15 625.59

2007:3 52.468 0.9 4.82

2,031 776 1,256 634.0 323.8

839.3 357.9 232.4 233.7 160.8

646.44 627.42 3.0 632.93 638.11 626.54

2007:4 52.483 0.8 4.75

1,881.3 1,833.3 1,908.7 1,904.3 1,845.5 1,403.1 1,351.6 1,433.5 1,412.2 1,380.1 478.1 481.7 475.1 492.1 465.3 78.1 74.1 74.6 75.7 75.8 196.16 198.59 200.26 200.07 200.74

2,373 964 1,409 627.7 313.5

1002.0 411.6 259.8 331.5 238.1

647.46 633.00 2.2 628.43 629.34 621.50

2006:4 52.065 1.0 5.83

1,830.2 1,367.9 462.3 75.2 201.83

1,913 776 1,137 633.1 331.4

1,734.7 1,261.1 473.6 71.4 192.68

1,715 697 1,018 621.0 327.1

909.2 439.0 139.7 369.1 140.8

621.70

627.02 859.7 385.9 167.8 275.0 180.1

655.03 632.25 3.5 628.00

2008:2 52.784 0.5 4.9

651.23 631.43 3.0 633.39

2008:1 52.362 0.0 4.9

1,589.8 1,139.7 450.2 68.0 189.20

1,604 652 951 608.7 319.7

659.3 350.6 109.9 206.2 269.5

616.104

655.71 627.87 4.1 622.20

2008:3 52.222 -0.5 4.0

1,551.5 1137.2 414.3 66.9 185.34

1,457 629 828 618.7 319.5

444.4 174.9 40.4 235.5 384.7

612.122

655.05 622.18 5.2 617.97

2008:4 51.810 -1.3 3.6

1,561.8 1124.2 437.5 66.0 171.18

1,124 505 620 581.9 300.9

681.5 326.0 75.2 226.6 169.6

607.956

647.07 603.94 6.9 613.70

2009:1 52.210 -0.3 2.1*

Dimensions of the Great Recession Not as severe as the Great Depression of the 1930s, the U.S. recession that started at the end of 2007 has proven to be more severe than almost every other one since then. Late in November 2008 the National Bureau of Economic Research (NBER) confirmed December 2007 as the peak of the last U.S. economic expansion. Shortly before that, in September 2008, a financial panic ensuing from the collapse of Lehman Brothers precipitated a more intense phase of the recession, until then more moderate in many dimensions. Hawaii data illlustrate much of what happened pre- and post-Lehman. The slowdown in Hawaii’s economic expansion had emerged some years earlier. Sales of existing homes peaked at the end of 2004, in seasonallyadjusted terms. Homebuilding peaked in 2005, and constant-dollar total visitor expenditure also peaked that year. The subsequent erosion in residential investment and in real tourism receipts during 2006 and 2007 began a process that ultimately dragged Hawaii into recession after

the financial crisis began in August 2007 and intensified in September 2008. Accompanied by loss of lift following the shutdown of Aloha Airlines and ATA in the spring of 2008, the tourism contraction deepened. The corresponding downward trend in Hawaii real retail sales in 2006 and 2007 steepened sharply in fourth quarter 2008 after the post-Lehman financial panic (see graph below). Hawaii’s recession was exacerbated in mid-2008 by the rise in global petroleum prices that peaked that summer, but subsequent declines also eased somewhat those pressures on Hawaii’s highly-petroleum dependent economy. The commodity price rebound that has raised oil prices to about half year-earlier levels in 2009 is a complcation for the stabilization that is starting to show sustainability in domestic tourism (facing page). The starting date picked by the NBER for the current recession corresponds with the start of job loss in Hawaii, which proceeded throughout most of 2008 and into 2009.

Hawaii real visitor expenditure

Hawaii real retail sales

(annual, billion 2008 dollars )

(quarterly, billion 2008 dollars at annual rates, s.a.)

16

32

15

30

14

28

13

26

12

24

Gulf 9/11 2008

11

2008Q4

22

NBER recessions shaded Jul90

10 85

Mar91

90

Mar01 Nov01

95

00

Dec07

05

20 10

98

00

02

04

06

Crude petroleum prices

Hawaii payroll employment

(monthly, U.S. dollars per barrel)

(monthly, thousands, seasonally-adjusted)

08

10

640

160

630 620

80

610 600

40

590 NBER recession shaded

NBER recessions shaded

580

20 00

6

01

02

03

04

05

06

07

08

09

2004

2005

2006

2007

2008

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

2009

Tourism recovery interrupted by H1N1-A virus Despite some weakness in March 2009--an unseasonably low volume of “Spring Break” domestic travel to Hawaii following two recent years with early Easters (2005; 2008)-domestic tourism had already begun to establish recession lows when the collapse of Lehman Brother was occurring back in September 2009. To be sure, it would have been hard for the average person to know what was going on. Monthly visitor arrivals are not seasonally-adjusted in the official statistics, nor was the stabilization in seasonallyadjusted domestic arrivals even clear until well into the winter of 2009. At the very moment in March 2009 that stock prices reached their low after the peak of October 2007, Hawaii domestic visitor arrivals were dropping out of the seasonally-adjusted plateau to which numbers had settled in the July 2008-February 2009 period, with the notable exception of September 2008. As April and May 2009 domestic visitor counts confirmed, however, seasonally-adjusted domestic volumes have been bouncing along a flat and volatile trajectory since second quarter 2008. Importantly, they are not declining. In the international segment, things are more complicated. As this report originally would have “gone to press” at the end of April 2009, the global H1N1-A viral pandemic emerged, casting doubt on fresh tourism forecast revisions. Now, with two months’ daily data on international passenger arrivals available, a “mini-SARS” pattern is emerging. International arrivals, after recovering from the Lehman event, are now falling because of this new biological event. The decrease appears to be less profound than SARS in 2003, but meaningful quantitatively. The hotel industry, which cannibalized its own utilization rate by raising hotel room rates more than 25 percent before the recession, has now been further unsettled by a small increase in the transient accommodation tax rate intended to help fill the State’s balanced-budget gap. Deep discounting has yet to stabilize hotel occupancy, although arrivals may finally have found a bottom.

Monthly domestic visitor arrivals (thousands, seasonally-adjusted, log scale)

600

Passengers Visitors

NBER recession shaded

550 500 June*

450 May

400 Aloha Airlines/ ATA shut down

Jul Sep

March

350 2004

2005

2006

2007

2008

2009

*based on daily data through June 25, 2009

Monthly international visitor arrivals (thousands, seasonally-adjusted, log scale)

190 180 170 160 150 140 Lehman

130 H1N1*

120

SARS

110 03

Hawaii real hotel performance indicators

05

06

07

08

09

*includes June estimate based on daily passenger arrivals data through June 25, 2009

(quarterly, percent of capacity (right scale), 2008$ per night (left scale), all seasonally-adjusted)

85

Occupancy (right) Real room rate (left)

04

Domestic and international visitor arrivals back to the late-1990s

80

(thousands, seasonally-adjusted, log scales)

75

500

NBER recessions shaded

450

70

210 200

65

190

60

180

400 350 300 250

170

200

160 NBER recessions shaded

150 96

98

00

02

04

06

08

Domestic (right scale) International (left)

9/11 Asian crisis

250 Y2K

150 Lehman

Data sources: Hawaii DBEDT for visitor counts (not seasonallyadjusted); hotel data from UHERO, as produced by Hospitality Advisors LLC; all seasonal adjustment and deflation by TZE

300

H1N1*

SARS

100 96

98

00

02

04

06

08

10

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

7

No lags this time Historically it has not been uncommon for Hawaii macroeconomic performance, by a variety of aggregate indicators’ experiences, to lag the mainland by months or quarters, and even years. During the S&L mortgage lending crisis of the 1980s, and the ensuing formation of the Resolution Trust Corporation (RTC), the cyclical trough in homebuilding nationwide coincided with the peak of the “Japan bubble” in Hawaii’s housing cycle around 1990. (Graph below; in the aggregate, statewide homebuilding peaked in Hawaii in 1990, but earlier on the neighbor islands than on Oahu.) The peak of the residential investment cycle coincided, this time, in the year 2005. Similarly, Hawaii’s unemployment rate moved widely “out of synch” with the national average during the early-1990s recession. While U.S. unemployment rates began rising almost immediately at the time of Iraq’s invasion of Kuwait

U.S. and Hawaii unemployment rates: cyclical movement aligned in the current recession (percent of labor force through May 2009)

9 U.S. Hawaii

8 7 6

9/11

5 4 3

NBER recessions shaded

2

U.S. and Hawaii residential construction cycles: peaks aligned in the current recession

90

(U.S. in million housing starts at annual rates, right scale; Hawaii in units authorized by building permit, left scale; both scales logarithmic)

U.S. starts (right scale) Oahu units (left scale) Neighbor islands (left)

3.2

1.6

0.8 1000

100

0.4

NBER recessions shaded

75

80

85

90

95

00

05

92

94

96

98

00

02

04

06

08

and the commencement of Operation Desert Shield in mid1990, Hawaii’s unemployment rate was actually at its cyclical low point in the middle of Operation Desert Storm during winter 1991. Hawaii unemployment peaked years after U.S. unemployment peaked in 1992. (Hawaii unemployment rates did not even rise during the “dot.com” recession of 2001 except for the increase associated with the 9/11 event in September 2001.) In the 2008-09 recession, however, not only have unemployment rate movements been highly synchronized between Hawaii and the mainland, home price movements that were a precursor to the broader economic downturn were equally synchronous around their peak in early-2006. As recently as the 1990s, home price movements had been highly unsynchronized between Hawaii and the mainland, with Hawaii lagging by years.

10

Comparison of median home price movements and Case-Shiller index movement Median single-family home prices

(thousand dollars, s.a., and 2000Q1 = 100; log scale)

(thousand dollars, s.a., log scale)

800

240 Orange Co., CA Maui Oahu

Peak: Synchronous

NBER recessions shaded

200

Oahu Maui

Peak: Synchronous

400 Oahu

160 Case-Shiller

120

Oahu (median) Maui (median) S&P Case C Shiller 20

Maui

Troughs: Asynchronous

O.C.

200 96

8

98

00

02

04

06

08

00

01

02

03

04

05

06

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

07

08

09

Hawaii’s low mortgage risk experience As has been true historically, but is especially distinctive in the current recession, Hawaii’s low mortgage delinquency rates set apart its economic performance from the national average. Broadly defined as 30 days or more past due, Hawaii’s mortgage delinquency rate is in the lowest quintile nationwide. Hawaii’s four counties ranks among the lowest nationwide as well in terms of serious delinquency, defined as 90days or more past due. Notably, these data include all residential mortgages with Hawaii collateral including those properties held by offshore investors. In many

cases, particularly in Neighbor Island resort communities, the latter comprise second-home purchases by mainland investors enabled by the same slackening of underwriting requirements that contributed to mortgage lending problems and the financial crisis generally. Correlated default risk, exacerbated by the simultaneity of the increase in mortgage delinquencies nationwide, contributed to the depreciation of mortgage-based CDOs (collateralized debt obligation) and related derivates such as credit default swaps (CDS) that precipitated the liquidity crisis.

Mortgage delinquency rates by state at end-2008 (30 days or more past due) Ranked by delinquency rate

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Mortgage delinquency rates continue to run lower in Hawaii than elsewhere, serious delinquency (90+ days) on Oahu is 1.6% compared to 12% in Dade Co. FL, 10% in Riverside Co. CA, and 8% in Clark Co. NV.

Hawaii 5.29% (90+ days 1.75%)

0

2

4

6

8

10

Mississippi Nevada Florida Michigan Georgia Indiana Louisiana Tennessee Alabama Ohio Arizona West Virginia California Texas Rhode Island

13.11 11.12 11.09 11.08 10.73 10.59 10.13 9.92 9.69 9.49 9.46 9.40 9.13 9.01 8.75

12

Percent 30 days or more past due Source: Mortgage Bankers Association

Mortgage delinquency rates by county at end-2008 (90 days or more past due) Darker shading indicates higher delinquency rate

Selected Counties: (percent of loans) _________________ Dade, FL 12.26 Merced, CA 10.57 Clark, NV 8.24 Maui, HI 3.15 Hawaii, HI 3.08 Honolulu, HI 1.64 _________________

Resilient Oahu home prices reflect underlying economic fundamentals, not imminent collapse

Source: Federal Reserve Bank of New York (image not geographically accurate)

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

9

Real Estate still receding Total sales of existing single-family homes and condominiums in Hawaii have declined from an annual pace around 21,000 statewide at the peak of the last cycle in 2005 to an annualized pace around 10,000 trades in the first several months of 2009. Low interest rates during winter 2009 helped surge a huge wave of refinancings as a deliberate outcome of Federal Reserve policy of “credit easing” or “quantitiative easing.” This was an explicit strategy of extraordinary purchases of long-dated U.S. Treasury securities designed to maintain low interest rates, combined with outright purchases of mortgage-backed securities and agency debt designed to reduce the risk premium in mortgage financing costs. By enabling large numbers of households to reset the terms of their mortgage liabilities, the hope was to help partially ofset much of the wealth loss associated with falling stock and housing prices by reducing debt burdens. Higher bank mortgage underwriting standards may have limited somewhat the efficacy of this strategy. At the same time, sales volumes in Hawaii--a “non-recourse” state with low mortgage delinquency and foreclosure volumes--did seem to respond positively to the low rate environment. Seasonally-adjusted sales volumes on Oahu rebounded through mid-2009 (graph at upper right). Greater absorption in the first few months of 2009 drew down months of inventory remaining on Oahu (graph at rigth). A sudden rise in longterm interest rates caused mortgage interest rates to rise beginning mid May 2009. Home prices on Oahu began fading more quickly in the early months of 2009, on a seasonally-adjusted basis. Oahu condominium prices looked somewhat more steady through mid-year. The drop in sales and rise in inventories that followed the postLehman financial panic did reverse itself in first half 2009, but whether sales have indeed found the bottom of the home sales recession that began at the end of 2004 remains an open question.

Oahu sales of existing homes (Monthly units at annual rates, seasonally adjusted, logarithmic scales)

800

Condominium Lehman

400

Single-family 200 NBER recession shaded

2005

2006

2007

2008

2009

Months of inventory remaining on Oahu (at existing sales rates and last month’s inventory)

14 12 10 8

Single-family

6

Lehman

4

Condominium

2 NBER recession shaded

0 2005

Oahu existing single-family home median sales prices

2007

2008

2009

Oahu existing condominium median sales prices (thousand dollars, seasonally-adjusted, level scale)

(thousand dollars, seasonally-adjusted, level scale)

700

350

600

300

500

250

400

200

300

150 NBER recessions shaded

NBER recessions shaded

100

200 00

10

2006

01

02

03

04

05

06

07

08

09

00

01

02

03

04

05

06

07

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

08

09

Residential investment: at root At the roots of the financial crisis, and even undelrying the explosion of sub-prime lending in the years preceeding the crisis that has left such a legacy of toxic mortgage-banking related assets, is a “garden variety” housing cycle. Residential investment is one of the components of gross domestic product that, though small, vary widely in a cycle in which the range from peak to trough is larger than virtually every other component of GDP. A tremendous amount of credit creation is associated with this residential investment cycle, because wealth in the form of home equity comprises the dominant form of household wealth, larger even than stock market exposure through household portfolio investment and retirement programs such as IRA and 401k plans. Financial innovation during the last decade extended the reach of securitization to a broader class of mortgage assets than those traditionally supported by conforming fixed-rate, 30 year mortgages sponsored by agencies like Fannie Mae and Freddie Mac. Sub-prime and Alt-A mortgages that had lower underwriting standards were securitized by private packagers, which increased the

financial system’s exposure to default risk as the housing cycle turned after 2005. As illustrated below, unlike the cycle in the 1980s and 1990s that preceded the post-2005 housing recession, the recent cycle was one in which U.S. and Hawaii homebuilding cycles were in phase. Both residential investment cycles peaked at around the same time in 2005, whereas in the 1980s/90s cycle the trough of the U.S cycle in 1990 coincided with the peak in the Hawaii cycle. At that time, the so-called Japan Bubble was having an unusually influential effect on Hawaii housing investment, while the U.S. mainland was working its way through the down years that followed the S&L crisis of the 1980s. Equally striking as the synchronous downturn in Hawaii in U.S. housing investment since 2005 is the difference between the extent to which U.S. homebuilding has fallen well below the lower bound of the range over which it has varied cyclically during the last half century. In Hawaii, where complete data aren’t available for as long, the current homebuilding downturn remains within the lower bounds of housing decreases from the last few cycles.

Hawaii housing authorizations by building permit

U.S. housing starts

(quarterly, units, seasonally-adjusted)

(monthly, million units at seasonally-adjusted annual rates)

4000

3200

2000

1600

1000

800

U.S. residential investment decline has broken out of the lower bound of the cyclical range of the last half century NBER recessions shaded

400

500 75

80

85

90

95

00

05

10

60

65

70

75

80

85

90

95

00

05

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

10

11

Notoriously cyclical, construction is problematic While the dominant view among economic forecasters is that a broad-based reversal of private U.S. real investment expenditure should mark the shift from negative real GDP growth in first half 2009 to positive and accelerating real GDP growth in second half 2009, that pattern of investment-led recovery may be uneven across segments. Gains in equipment and software investment, and even consumer durables purchases, seem likely. Residential investment has overshot volumes consistent with mere maintenance of the per capita housing stock in the U.S., but here as well as in nonresidential investment the constraints imposed on the overall economy by the credit crunch in U.S. commercial banking will be slow to recede. Moreover, an overhang of commercial mortgage refinancings expected to surge in the next two to seven years poses additional uncertainties in spite of efforts by the Federal Reserve to directly restore channels of securitization for commercial mortgages. In Hawaii these issues also weigh on the construction outlook. The most recent forecast of the University of Hawaii Economic Research Organization (UHERO), published in March 2009, anticipated overall construction recovery in Hawaii no earlier than 2011, partly as a result of implementation lags associated with federal fiscal stimulus. A quarterly forecast prepared by the author for the January 2009 meeting of the Hawaii Council on Revenues, before passage of the American Recovery and Reinvestment Act of 2009 (ARRA: federal fiscal stimulus) envisioned a similar timetable for construction recovery, with less rapid recovery for real construction spending. Somewhat surprisingly, at least some observers appear to have expected that federal fiscal stimulus would have strong near-term impacts, even though it is virtual textbook doctrine that implementation lags for fiscal policy are long and variable. Material positive impacts of the 2009 federal stimulus initiative, comprising at three percentage points of GDP an injection twice the size of the New Deal in the 1930s, should be largest in second half 2009 and first half 2010, gradually tapering off over the four quarters ending in mid-2011.

Hawaii private construction commitments: monthly totals and cyclical (trend) component (monthly, million 2008 dollars, s.a., log scale)

400

200

100 NBER recessions shaded

65

70

75

80

85

90

95

00

05

10

Hawaii public construction commitments (annual, million 2008 dollars, s.a., log scale)

3200

“Burns”

“Waihee”

Includes county, state and federal government contracts

1600

800 DBEDT data Old BOH data UHERO 3/09 forecast 400 65

Brewbaker construction forecast for January 2009 meeting of Hawaii Council on Revenues

70

75

80

85

90

95

00

05

10

15

UHERO Hawaii construction forecasts (annual, billion 2008 dollars, log scale)

(quarterly, million 2008 dollars, s.a., log scale)

8 2000

7 6 5

1000 Actual real contracting H-P trend Forecast 500 80

12

85

NBER recessions shaded

NBER recessions shaded

Strikes

90

95

00

05

10

Mar 2007 Sep 2008 Mar 2009

4

3 15

80

85

90

95

00

05

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

10

15

What a long, strange trip it has been When the Chair of the National Bureau of Economic Research (NBER), Stanford University Professor Robert Hall, opined to a conference of the National Association for Business Economics (NABE) in October 2008 that the jury was still out whether recession should officially be declared (conceding its likelihood), U.S. real GDP still hadn’t experience a sustained and broad pattern of decrease that would be consistent with economy-wide recession. The intensity of the financial panic in the fall of 2008 that followed the collapse of Lehman Brother in September, despite extraordinary, and globally coordinated, monetary policy interventions in the following months, helps explain why only one month after the NABE meetings the NBER officially dated the peak of the last U.S. expansion as December 2007. Clearly, the implosion of U.S. consumer spending in second half 2008, much of which accompanied the stock market collapse in the fall’s

financial panic, served as the decisive accelerant. After December 2007, widespread job loss averaging 150,000 to 200,000 per month through September 2008 contributed to a creeping deterioration in consumer sentiment. A collapse of real consumption expenditure in third and fourth quarter 2008, marked by an unprecedented jump in the U.S. personal savings rate, dragged real GDP into the vortex. By end-2009 it appears increasingly likely that the U.S. personal savings rate will return to its historic range of 8-10 percent (from the 1960s-90s). As of mid-2009, total nominal U.S. personal savings had in one year reversed two decades of decline to exceed personal savings almost a quarter century ago. While investment is beginning to mobilize in anticipation of a consumption recovery, and massive inventory reductions now presage a production recovery, it seems quite plausible that a significant, secular shift in consumer behavior may have been initiated in 2008.

U.S. personal savings rate

U.S. real disposable income and consumption

(percent of disposable income, through May 2009)

(monthly, trillion 2008$ at annual rates, s.a., log scales)

9.2 Disposable income (right) Personal consumption (left)

10

9.0

8

8.8

6

8.6

4

8.4

2

8.4 8.3 8.2 8.2 8.1 8.0

0 -2

NBER recession shaded

NBER recessions shaded

7.9 2006

2007

2008

-4

2009

92

94

96

98

00

02

04

06

08

Quarterly U.S. real aggregate consumption expenditure growth through first quarter 2009

Quarterly U.S. real GDP growth through first quarter 2009

(quarterly percent change at annual rates)

(quarterly percent change at annual rates)

6

6

4

4 2

2

0 0 -2 -2 -4

-4

NBER recession start date 2007.4 and presumed 2009.3 end shaded

-6

-6

-8 05

06

07

08

09

05

06

07

08

09

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

13

Unwinding the monetary intervention Combined with an historically unprecedented fiscal policy intervention, the extension of monetary policy intervention beyond traditional use of interest rate targeting is one of the hallmarks of the current recession. When the federal funds rate, reached zero at the end of 2008 a new policy approach was needed. Implemented gradually at first in fourth quarter 2008 and then expansively in early-2009, this was executed especially through securities acquisition, while other channels of intervention actually diminished (graph at right). The Fed’s move to what Chairman Bernanke has called “credit easing,” or quantitative easing in the theoretical literature, is exhibited in the massive increase in the Fed’s balance sheet as depicted. Because this injection coincided with what the Depression-era economist J.M. Keynes called an increase in “liquidity preference” by households and businesses following the collapse of Lehman Brothers, the Fed was essentially acting to prevent the collapse of the money supply that

would otherwise have accompanied the flight towards the save haven of bank deposits and cash equivalents. This increase in the demand for money required a countervailing increase in its supply. In the short-run, such behaviors are rational responses to perceptions of increased risk. In the long-run, as behavior returns to normal and households and businesses go back to minimize holdings of liquidity that forego higher interest earnings and other returns (including utility from consumption) the Fed will gradually have to reverse its massive monetary injection to avoid becoming a source of future price inflation.

Federal Reserve assets (billion dollars, weekly through mid-June 2009) 2500 Securities Other assets Term Auction credit Commercial Paper funding Central bank swaps

2000

NABE real GDP forecasts: 2nd half recovery (quarterly percent change in real GDP at annual rates)

1500

6 4

1000 02/08

2

05/08

0

500

10/08 NABE 05/09

-2 11/08

Actual May 09 forecast

-4

0 Aug-07

Dec-07

Apr-08

Aug-08

Dec-08

Apr-09

02/09

-6

Federal Funds rates under the Taylor Rule (percent, actual and under NABE forecast assumptions)

NBER recession shaded

-8 06

07

08

09

10

U.S consumer price inflation rates show no sign of future inflation risk, but only time will tell (percent changes, year-over-year)

NBER recessions shaded

6

4

Headline Core

5

8

2

4

0

3 2

Fed funds rate Taylor Rule

-2 98

00

02

04

06

08

10

12

1

Target Fed Funds as f[inflation gap, output (growth) gap]: r* = [4.5 + (0.5)(p - p*) + (0.5)(y - y*)]

0 -1 03

14

04

05

06

07

08

09

r p y

= Fed Funds rate = increase in the core CPI [p* =2 (target)] = real GDP growth rate [y* = potential GDP growth]

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

Nominal U.S. Treasury yields (percent, adjusted to constant maturities)

Jul 06

5

Jul 07 4 June 2009

Jul 08 Apr 09

3 Dec 08 2

1

-y r 30

-y r 20

-y r 10

7yr

5yr

0 FF 1y 2- r y 3- r yr

Monetary policy and beyond: fiscal policy An intensification of the global financial crisis in September 2008 abated in winter 2009 with a stabilization of equity valuations, enactment of federal fiscal policy stimulus, and signs going into the spring that economic indicators might exhibit a transition from recession to stabilization during the course of the year. Indeed, much talk of “green shoots” of spring accompanied a rebound in stock prices during the latter two-thirds of March through April and May, before running into the “lawn mower” of summer by June 2009. At the end of second quarter 2009 long-term yields on U.S. Treasury notes and bonds had moved up sharply at the tail end of the stock price recovery. Concern about the persistence of federal budget deficits implied by the fiscal stimulus plan and other ambitious initiatives including health care systemic reform began to raise two kinds of concerns both unfavorable for bond prices. First, under a scenario in which long-term inflation expectations remained “well-anchored,” real long term interest rates might have to rise to induce the necessary global savings required to finance persistent deficits. Second, any erosion of monetary policy credibility might lead to higher future inflation than such deficits might imply if the central bank’s commitment to restraining future inflation was not called into question. Neither of these outcomes is necessary in the sense that a stable inflation environment and higher real interest rates could exist in a formation that doesn’t unduly crowd out private investment under a sufficiently robust economic recovery. At the end of second quarter 2009, long-term inflation expectations remained contained.

Real U.S. Treasury (TIPS) yields (percent, adjusted to constant maturities)

3

Long-term inflation expectation implied by the difference between nominal and real yields (percent, includes a latent (unobservable) inflation risk premium and liquidity risk premium)

Jul 2007 Jul 2006 2

3 Jul 06, 07, 08 2

Jun 2009

Mar 2008

1 Jul 2008

Jun 2009 1

Mar 2008

Feb 2009

30

-y r

-y r 20

-y r 10

7yr

FF 1y 2- r yr 3yr

0

5yr

0

-1 Nov 2008

30 -y r

20 -y r

yr 10 -

yr 7-

yr 5-

1y 2- r yr 3yr

FF

-2

Bank of Hawaii Economic Research Center https://www.boh.com/econ/512_539.asp (June 30, 2009)

15