THE COLBY ECONOMIC OUTLOOK FOR THE MAINE ECONOMY

by Michael R. Donihue, Associate Professor of Economics, and Stephen Drunsic, Senior-Seminar Economics student, both at Colby College

Last December, the Colby Economic Outlook (CEO) celebrated its 7th year of providing short-term projections for the U.S. macroeconomy and the state of Maine. Each year, students in the senior seminar in economic forecasting study methods for data analysis and forecasting using the Colby Quarterly Econometric Model of the U.S. economy under the direction of Michael Donihue, Associate Professor of Economics. This year's CEO highlights the impact of the rebenchmarking of the National Income and Product Accounts on the Colby Model and the development of the Colby Coincident Indicator of the Maine Economy. Downloadable copies of the CEO are available on the internet at http://www.colby.edu/economics/ceo.

From our perspective, 1996 was somewhat of a puzzle for the Maine economy. For much of the year the construction industry in central and southern Maine was in very good shape, and anecdotal reports on retail sales and income throughout the year indicated that the economy was on the upswing. Yet the official data on employment painted a very different picture throughout the year.

Total nonagricultural employment in the state declined in five of the first seven months of the year, with employment in July lower than it had been in over a year. The last time employment declined for two consecutive quarters was in the early days of the recovery from the 1990-91 recession.

Employment in the third quarter increased at an annual rate of 0.7%, and based on that information we estimated in early December that employment growth during 1996 would average just 0.4% -- an estimate confirmed by data released in February 1997 from the State Planning Office. We predict that total employment will grow by less than 1% throughout the forecast horizon as Maine continues to face a loss of manufacturing jobs. We also predict that the unemployment rate will remain roughly constant throughout the forecast horizon at about 5.1%.

Looking across other sectors of the Maine economy:

• Personal income is forecast to grow by 4.3% in 1996, 4.9% in 1997, and 5.3% in 1998. In real terms, this translates into income growth of 2.2% for 1996, 2.4% for 1997, and 2.6% in 1998 given Colby's national forecasts of consumer prices.
• Consumer retail sales for December have not yet been reported, but our estimate is that retail sales grew by 5.1% in 1996. We believe that retail sales growth will slow noticeably in 1997 to 3.5%, primarily as a result of the sluggish growth in employment.
• Lodging sales, a new feature of the Maine model this year, are forecast to grow by 4.5% in 1996, and jump by 7.9% next year, despite a predicted increase in the Canadian/U.S. dollar exchange rate. Much of the uptick in growth next year is forecast to occur because of strong disposable income growth nation-wide during 1996.
• Commercial electricity sales, also new to the Maine model, are forecast to rise by 2.3% in 1996, 1.8% in 1997, and 1.4% in 1998. We believe that this variable provides a useful alternative measure of the health of the commercial sector of the Maine economy.
• Finally, total undedicated revenues in the state are forecast to end Fiscal 1997 at $1.87 billion. Fiscal 1998 should see revenues increase to $1.99 billion, according to our projections. Given the $1.81 billion in appropriations for FY '97, and Governor King's proposed budget for FY '98 of $1.82 billion, Maine should realize a small budget surplus in both periods according to our projections.

A New Coincident Indicator For the Maine Economy

Diagnosing the overall health of the Maine economy proved to be a daunting task for this year's seminar. Unlike the national economy, whose behavior can be pegged closely to GDP, business cycles in Maine frequently appear to be very different depending on which indicator you use as a guidepost. Furthermore, business cycles in Maine tend to vary greatly from the national cycles both in terms of duration and magnitude which means that diagnostic variables which work well at the national level may be inappropriate at the state level. Dealing with regional economic data has always been difficult because of inherently different behavioral dynamics and the limitations of the scope and timeliness of the variables. Gross State Product data, for example, are only reported annually, and then with a lag of up to 3 years, rendering it useless as a contemporaneous indicator of the health of the Maine economy.

Two widely published aggregate indicators of the Maine economy, the Maine Business Index, published by the Center for Business and Economic Research, and the Maine Economic Growth Index, published by the State Planning Office, provide useful insights into business cycles in Maine, but like all summary measures have some limitations. The Maine Business Index, because it is constructed from eleven different variables, is reported with a lag of up to four months. The Maine Economic Growth Index, released with a one-month lag, appears at times to overstate the health of Maine's economy due to a heavy reliance on service-sector employment and consumer retail sales. Thus, during 1995 the seminar began a research task of developing a coincident indicator of our own, which would have some of the same historical performance features as do these other indicators, but could coincide with the release of the CEO and include a short-term forecast.

The Colby Coincident Indicator of the Maine Economy (CCI) was constructed using a methodology similar to that developed by the National Bureau of Economic Research. The CCI consists of four components: new housing permits (lagged one period), commercial electricity sales, personal income, and nonmanufacturing employment. These were chosen on the basis of their ability to track economic activity in a consistent manner. All components are available at roughly the same time, facilitating a timely release of the CCI. The CCI begins tracking the economy in the first quarter of 1976 and, as an added feature, includes a two-year projection into 1998 using forecasted data.

According to the CCI, economic activity in Maine peaked during the second quarter of 1984 -- signaling a quick recovery from the recession of 1981-82. Unlike the Maine Economic Growth Index, Gross State Product, and the Maine Business Index, the CCI does not show continued growth throughout the rest of the 1980s. We interpret the leveling off in the CCI at relatively high levels of economic activity during the latter half of the 1980s as reflecting a period of readjustment and restructuring of Maine's industries which predates the national recession of 1990-91.

Entering the 1990s, the CCI drops precipitously into the trough of the recession in the first quarter of 1991, mirroring the national recession. Unlike the national economy, however, the recovery in Maine has taken much longer. As of the second quarter of 1996, the CCI, with a value of 99.38 is still well below the 1987 base-year level of 100 -- an indication that the economy has yet to attain pre-recession levels. Unlike the coincident indicators mentioned above, which report continued growth above pre-recession levels through 1996, we believe that the sluggishness of the CCI in the mid-1990s more accurately reflects a slowing pace in the economy. According to our projections, we somewhat pessimistically predict more of the same through 1998.