ISSUE 6 – 1995FEDERAL RESERVE BANK OF DALLAS

From Crude OilTo Computer

Chips

T

“I think there is a world market formaybe five computers.”

—Thomas Watson,chairman of IBM, 1943

“I have traveled the length and breadthof this country and talked with the bestpeople, and I can assure you that dataprocessing is a fad that won’t last outthe year.”

—The editor in charge ofbusiness books forPrentice Hall, 1957

“There is no reason anyone would wanta computer in their home.”

—Ken Olson, president,chairman and founder ofDigital Equipment Corp., 1977

echnological innovations arerapidly changing the way people

live. Computers, fax machines,mobile phones and online computerservices allow people to work fasterand more efficiently, with ever-expanding access to information.With each new generation of ma-chine, high-tech products becomemore pervasive household fixtures.No longer is it only the wealthywho can afford to own a VCR orcellular phone. New products arebeing developed every day to meetconsumers’ voracious appetitesfor faster, more efficient ways toconduct business and enhanceleisure time.

Thanks to the acceptance andaffordability of new technologies,certain high-tech industries havegrown into a rapidly expandingsegment of the economy. In Texas,employment at high-tech firms hasgrown twice as fast as the state’soverall economy during the past10 years.1

Although Texas’ economic rootsare grounded in agriculture and oil,the state has pioneered many tech-nological innovations. Dallas’ TexasInstruments (TI) was among thefirst companies to mass producetransistors, and a TI engineer, JackKilby, developed the integratedcircuit. Another Dallas company—Electronic Data Systems Corp.

(EDS)—was among the first firmsto offer data processing services.And Dell Computer Corp., head-quartered in Austin, is the fifthlargest maker of personal computersin the world and is one of thefastest growing U.S. corporations.

Despite recent lean years for thedefense industry, Texas defensegiants—including LTV, Bell Heli-copter and Lockheed (formerlyGeneral Dynamics)—have beenimportant catalysts for high-techadvancement. As such firms movedto Texas during the World War IIbuildup, they brought scientists andengineers. Other workers becameskilled in electronics, telecommuni-cations, and weapons and aerospacemanufacturing. Now, even as em-ployment in the defense industrywanes, the private sector is makinguse of defense-related technologicaladvances.

How Much High-Tech in Texas?

The Texas economy, once drivenby resource-based industries such asfarming, ranching and oil produc-tion, is evolvinginto a moreknowledge-based economy.While the oiland gas extrac-tion business is

How TechnologyIs Changing theTexas Economy

I N S I D E

The Changing MeaningOf Money

—

A Look at the Top U.S.Trading Partners

still very important to the state’seconomy, its share of total employ-ment fell from its early 1980s peakof 5 percent to about 2 percent in1994.2 In contrast, the share ofTexas employment in high-techindustries rose from about 2 per-cent in the mid-1970s to 3.4 percentin 1994.3

As Chart 1 shows, high-techemployment grew more than twiceas fast in Texas than in the nationas a whole during 1988–94.4 Texas’strongest performance relative tothe nation’s has come in computer-and telecommunications-relatedindustries.5 This category includesfirms that make computers, com-puter chips and cellular phones andfirms that provide programming ordata processing services. Since 1988,employment in Texas computer-and telecommunications-relatedindustries has grown more than eighttimes the national rate. Currently,the share of computer-related andtelecommunications-related em-ployment to total employment is2.5 percent in Texas and 2.1 per-cent in the United States.

Chart 2 shows the 10 largest high-tech industries in Texas. Texas’employment in four of these 10 in-dustries exceeds the national aver-age. These industries (highlighted

in bright red) are computer-relatedservices, electronic componentsmanufacturing, computer manufac-turing and communications equip-ment manufacturing—all of whichare computer- and telecommuni-cations-related industries.Computer-related services. As com-puter and communications technol-ogy has become more widely used,

the number of service firms has sky-rocketed. Thus, it is not surprisingthat the largest high-tech industryin Texas is service-related. Employ-ment in Texas’ computer-relatedservices industry has grown almost50 percent since 1988, slightly fasterthan the national average. Includedin this category are firms that pro-vide computer programming, dataprocessing, software design, systemsdesign and information retrieval.Plano’s EDS, for example, is thenation’s largest provider of com-puter services to business andgovernment.

The computer-related services in-dustry accounts for almost 1 percentof total Texas employment, whichis roughly the same size as Texas’fabricated metals manufacturingindustry. However, employmentstatistics may drastically underesti-mate the actual number of computer-related jobs in the Texas economy.Employment statistics count onlyfirms that produce a service, suchas programming. Programmers whowork for a bank, for instance, arenot counted. Because many firmsemploy their own programmers orsystems specialists, total employ-

Chart 1Employment Growth, 1988– 94

Percentage change

–40

–30

–20

–10

0

10

20

30

40

50

60

Total nonagriculturalAmmunitions and aerospace

Computer- and telecommunications-related

High-tech

United States

Texas

SOURCE: Bureau of Labor Statistics.

Chart 2High-Tech Employment in Texas(Shares of 1994 Total Nonagricultural EmploymentFor Largest 10 High-Tech Industries in the State)

Percentage of total nonagricultural employment

United States

Texas

0

.1

.2

.3

.4

.5

.6

.7

.8

.9

1

Computer-related services

Electronic components

manufacturing

Computer manufacturing

Research and development

Communications equipment

manufacturing

Medical instruments

manufacturing

Measuring and controlling

instruments manufacturing

Miscellaneous electrical

machinery manufacturing

Pharmaceuticals and drug

manufacturing

Electrical industrial apparatus

manufacturing

SOURCE: Bureau of Labor Statistics.

2

ment in this sector is probably muchhigher than the numbers suggest.

The greatest job growth in Texas’computer-related services industryhas been from firms that providesoftware design and computer pro-gramming (Chart 3 ). Since 1988,jobs in software production havegrown by 105 percent in Texas,compared with 76 percent at thenational level. Austin alone hasmore than 500 software companies.Computer programming employ-ment in Texas has risen 96 percentsince 1988, compared with 76 per-cent nationally.Electronic components manufacturing.As Chart 2 indicates, the secondlargest high-tech industry in Texasis electronic components manufac-turing. This industry includes firmsthat produce computer chips andcircuit boards, both hot commodi-ties worldwide. Since 1988, circuitboard manufacturing employmentin Texas has expanded four timesfaster than in the nation.

Several large companies pro-duce computer chips in Texas, in-cluding TI, Motorola, AdvancedMicro Devices (AMD), Hitachi,Cyrix and National Semiconductor.Nationally, the computer chip indus-try has undergone a retrenchmentin the last several years. However,

announced expansions by Hitachi,Motorola and TI and a worldwideshortage of computer chips suggestthat this segment of Texas’ high-tech industry has recovered andshould see strong growth in thefuture.Computer manufacturing. Texas isquickly becoming synonymous withcomputer production. The LoneStar State is home to Dell, Compaq,TI and AST. Employment at com-puter makers, while falling nation-ally, has risen strongly in Texas.6

Since 1988, Texas employment atcomputer makers has risen 34percent, and employment in thecomputer peripherals industry,which includes printers, has risenby 15 percent. Dell exemplifies thisgrowth; Dell recently announcedthat it would build a third facilityin Austin because its second facilitywill be at capacity when it comesonline in November.Communications equipment manufac-turing. Communications equipmentmanufacturing includes firms thatproduce telephone, radio and tele-vision equipment. Texas job growthin this industry has not been asstrong as in other high-tech sectors,mainly because new technology hasmade workers more productive.Nevertheless, employment growth

Chart 3Employment Growth in Computer-Related Services, 1988– 94

Percentage change

United States

Texas

0

20

40

60

80

100

120

Software designComputer programming

Computer systems design

Information retrievalData processing

SOURCE: Bureau of Labor Statistics.

in this industry has been positivein Texas while declining at thenational level. The Dallas/Fort Wortharea is the heart of Texas’ telecom-munications industry, with firmssuch as Nortel, DSC, MCI and Nokia.Recently, Dallas/Fort Worth waschosen as the site for the headquar-ters of PCS PrimeCo, a joint ventureof Bell Atlantic, Nynex, US Westand AirTouch Communications.

Where Are the Jobs?

Like Dallas, with its communica-tions nexus, other Texas cities haveattracted concentrations of high-tech industries. Computer-relatedservices and the production ofelectronic components, computersand communications equipment—major players in the Texas high-tech sector—are located mostly inmajor cities, as shown in Chart 4.7

Austin’s computer manufacturingsector provides more than half ofthe state’s jobs in that industry. Thecapital city also has a large concen-tration of computer chip makers,such as Motorola and AMD, andtheir suppliers, such as AppliedMaterials and Tokyo ElectronAmerica. As a result, one-fifth ofthe state’s electronic componentmanufacturing is in Austin.

With 52 percent of the state’stotal, Dallas/Fort Worth has the lion’s

3

Chart 4Where Texas’ High-Tech Jobs Are

Houston 17%

Austin 20%

Dallas/Fort Worth 52%

Rest of the state 11%

NOTE: Percentages are each city’s share of Texas jobs infour leading high-tech industries–computer-relatedservices, electronic components manufacturing,computer manufacturing and communicationsmanufacturing. The share of total employmentthese four industries contribute to the Texaseconomy exceeds the national average.

SOURCES: Texas Employment Commission and TexasState Comptroller of Public Accounts.

share of high-tech jobs. Over halfof the state’s computer-relatedservices providers, such as pro-gramming and software designfirms, are in D/FW. The area alsohouses most of the state’s commu-nications equipment manufacturingfirms, with almost 80 percent ofTexas jobs in that sector. D/FWtelecommunication firms, such asDSC, Siemens, Motorola and Erics-son, produce products ranging fromswitching devices used to transmitdata and voices to cellular phones.Like Austin, Dallas/Fort Worth hasa large concentration of computerchip manufacturers and is home tomore than 50 percent of the state’selectronic equipment jobs.

Houston’s strongest high-techindustries are computer manufac-turing and computer-related services.Home of Compaq Computer, Hous-ton has 17 percent of the state’shigh-tech jobs—rivaling Austin’s20-percent share.

Why Texas?

Texas’ history in the defense andoil industries helps explain whythe state has become a high-techmecca. Texas Instruments, for ex-ample, built on its success in the oilbusiness to become a large defensecontractor and later one of thelargest computer chip producers inthe country.8 But other factors havecontributed to the state’s appeal tohigh-tech firms as well.

Texas’ low costs, high-tech re-search, large labor pool and promi-nence as a worldwide distributionhub have drawn firms to the state.Many of Texas’ high-tech exportsgo to Mexico. In 1994, for the firsttime since state export figuresbecame available in 1987, Texas’leading export industry was elec-tronic components manufacturing,which contributed $11.2 billion instate exports. The industrial machin-ery and computer manufacturingindustry was a close second with$11.1 billion in exports. Of theelectronic components exports,$5.8 billion went to Mexico, along

with $2.4 billion of the industrialmachinery and computer equip-ment. Several electronics firms,such as General Electric, Toshibaand Philips Consumer Electronics,are located in the El Paso/Juarezarea and take advantage of themaquiladora program betweenthe United States and Mexico.

In addition, Texas is an ideallocation for firms exporting else-where. Several high-tech firms(such as Nokia, Zenith Electronicsand GWS Perlos—a phone partssupplier) have located manufac-turing plants or distribution centersat Alliance Airport in Fort Worth,partly because the airport’s centrallocation and air, rail and highwayaccess make it ideal for globaldistribution.

High-tech firms tend to clusternear one another to be close tosuppliers and skilled workers. Morethan 20 suppliers followed AppliedMaterials to Austin after its move in1988, for example. Texas’ laborforce is younger and faster growingthan the national average. And,while Texas’ level of educationalattainment is about even with thenational average, the skill distribu-tion is widespread.9 Despite its largepercentage of high school drop-outs, Texas also has a high percent-age of skilled workers who helpattract high-tech firms to the state.

High-tech companies have citedlow costs as another factor drawingthem to Texas. Although the realestate market has been improving inrecent years, Texas apartment rentsand construction costs are muchlower than the national average.The average price of a Dallas home,for example, remains about 10 per-cent below the national average.Texas is also a low-tax state. Amongthe 50 states, Texas ranks 31st inper capita state and local tax reve-nues and 42nd in per capita ex-penditures.

Finally, Texas has industry con-sortiums and universities that pro-vide high-tech research to benefithigh-tech industries. For example,Austin is home to two of the nation’s

4

“In Texas, employment

at high-tech firms has

grown twice as fast as

the state’s overall

economy during the

past 10 years.”

premiere research consortiums:Microelectronics and ComputerTechnology Corp. (MCC) andSematech. These consortiums enablecompanies with common require-ments for new technology to sharethe costs and risks of development.Also, the Technology LicensingOffice at Texas A&M University andthe IC2 Institute at the University ofTexas at Austin provide universityresearch that benefits high-techindustries.

Clouds on a Bright Future?

Although Texas’ high-tech futurelooks bright, a few clouds on thehorizon could impede employmentgrowth. Environmental consider-ations, such as water purity, coulddeter companies from relocating toor expanding operations in Texas.Water is an important input in thecomputer chip manufacturing pro-cess, and companies in Austin areconcerned about the water availa-bility from the Edwards Aquifer.

A lack of office space in someprime high-tech office districts isanother consideration. The officevacancy rate in Northwest Austin—the most popular area among high-tech companies—is about 3 percent.Even after completion of construc-tion projects under way, space maynot be available to meet demand.Although they are lower than thenational average, Dallas’ suburbanoffice rents have risen rapidly in thepast two years and are eating awayat one of Texas’ biggest draws.10

Signs also indicate that Texas’skilled labor market is tightening.Industry contacts in Austin reportthat they must look beyond Texas’borders to find skilled workers. Infact, the estimated unemploymentrate for engineers and softwaredevelopers is below 1 percent inAustin. Several high-tech compa-nies that recently located outside ofTexas cited the state’s tighteninglabor pool as a major factor in theirdecision. So far, Austin appears tobe the only Texas city straining atthe seams, but labor market pressure

could occur in other Texas citieswith a high concentration of high-tech industries.

Despite these obstacles, Texasshould be a major benefactor inthe quest for faster, more efficientways to work and better leisureproducts. High employment growthshould continue in high-tech in-dustries concentrated in Texas—namely, computer-related services,electronic components manufactur-ing, computer manufacturing andcommunications equipment manu-facturing—for several reasons.Industry consortiums, such as MCCand Sematech, and ongoing researchat Texas universities yield synergiesfor high-tech businesses. The state’salready strong base of high-techcompanies and suppliers can enticeother companies to relocate toTexas. And Texas’ growing laborforce and the ease of relocatingworkers from other areas of thecountry are valuable assets thatshould continue to attract high-techrelocations and expansions.

—D’Ann M. PetersenMichelle Thomas

Notes

1 In this article, we define high-tech toinclude the following three-digit Stan-dard Industrial Code (SIC) categories:pharmaceuticals and drugs, computermanufacturing, electrical transmissionand distribution equipment manufactur-ing, electrical industrial apparatus manu-facturing, household audio and videoequipment manufacturing, communica-tions equipment manufacturing, elec-tronic components manufacturing,miscellaneous electrical machinery manu-facturing, measuring and controllinginstruments manufacturing, photographicequipment and supplies manufacturing,computer-related services, and researchand development. Our definition ofhigh-tech industries is taken from theTexas Comptroller of Public Accounts.The comptroller’s office bases its defi-nition of high-tech on the followingcharacteristics: (1) employing a higherpercentage of technicians, engineersand scientists than most manufacturersand (2) having an above-averageresearch and development component.Because of recent budget cuts andmilitary personnel cuts, ammunitions

5

and aerospace industries are excludedfrom this analysis.

2 The oil and gas extraction industryaccounts for a larger share of state out-put than of total state employment. In1982, oil and gas extraction accountedfor roughly 18 percent of total stateoutput, compared with about 7 percentin 1994.

3 We use the Bureau of Labor StatisticsES202 employment data for the U.S. andTexas three- and four-digit SIC sectors.In the calculations of employmentshares, we use U.S. and Texas totalnonagricultural employment from theBureau of Labor Statistics EstablishmentSurvey. For the city employment data, weuse ES202 employment data providedby the Texas Employment Commission(TEC) and the Texas Comptroller ofPublic Accounts.

4 We use 1988 as our reference pointbecause several four-digit SIC sectors inhigh-tech industries were not availablebefore 1988.

5 Computer- and telecommunications-related employment is a subset of high-tech employment and includes computermanufacturing, electrical transmissionsand distribution equipment manufactur-ing, household audio and video equip-ment manufacturing, communicationsequipment manufacturing, electroniccomponents manufacturing, miscella-neous electrical machinery manufactur-ing and computer-related services.

6 Nationally, employment in computermanufacturing has fallen since 1988, butproduction has risen 136 percent. Overthe past two decades, computer manu-facturing has become much less labor-intensive because of new equipmenttechnology. While output has risensharply, new technology has enabledworkers to become more productive.

7 Due to confidentiality concerns, TECwill not release computer manufacturingemployment for Houston and FortWorth. We approximate this employ-ment by using estimates for the numberof jobs at computer manufacturersCompaq (Houston), Tandy Electronics(Fort Worth) and AST (Fort Worth).

8 See “Industry: High Tech and Defense,”Forces of Change (Austin: Texas Comp-troller of Public Accounts), 1994.

9 See Stephen P. A. Brown and Lea An-derson, “The Future of the SouthwestEconomy,” Southwest Economy, Novem-ber 1988.

10 While downtown Dallas has one of thenation’s highest office vacancy rates,the suburban rate has tightened. Mosthigh-tech companies are located in thesuburbs.

The ChangingMeaning

Of Money

rates yields two important relation-ships:

and

where nominal GDP growth equalsgrowth in the dollar volume ofgross domestic production (outputgrowth plus inflation). U.S. outputtypically grows at about 2.5 percentannually. Thus, the equation ofexchange strongly suggests that,over the long run, inflation can bekept at zero by limiting moneysupply growth to equal 2.5 percentminus growth in velocity.

Money holdings typically falland velocity rises as the spreadbetween a riskless short-term marketinterest rate and the average yieldon monetary assets rises. Thestability of the relationship betweeninterest rates and velocity is whatmakes it possible for money to bea useful indicator of not only in-flation, but also of nominal GDP(P × Y ), since GDP data are avail-able after a long lag, unlike data onmoney and interest rates. If velocityis predictable, then by controllingmoney supply growth, the FederalReserve can control long-run in-flation. While this sounds easy,shifts in how people conduct theirfinances and how they pay forgoods can undermine the stability ofthe money–GDP relationship, thusmaking the Fed’s inflation-fightingjob more difficult in practice.

History bears this out. The M1monetary aggregate that measuresthe money supply as checkingdeposits plus currency was oncetouted as the “holy grail” by mone-tarists. But M1 began to fall fromgrace in the mid-1970s when itsvelocity was unusually high, and M1growth underpredicted real GDP,based on prior velocity behavior.Then in the early 1980s, the interest-rate sensitivity of M1 jumped asfinancial innovations and deregula-

tion created new deposits thatcombined savings and transactionsfeatures and helped firms avoidholding non-interest-bearing de-mand deposits. As a result, atten-tion turned to M2, a broader andless interest-rate-sensitive aggregatethat was created in 1980.

M2 was redefined to include notonly conventional M1, passbooksavings accounts and small timedeposits, but also new types ofmoney, such as money marketmutual funds, overnight instrumentsand, in 1982, money market depositaccounts. M2 had a stable relation-ship with nominal GDP during the1980s (Small and Porter 1989).However, this relationship brokedown in the 1990s as M2 becamemore sensitive to bond yields andas households shifted toward bondand stock mutual funds and towardTreasury securities (see Duca 1995bfor references).

Such breakdowns in the link be-tween money and nominal outputhave spurred efforts to either rede-fine money to include new types of“money” or revise money modelsto account for changing relation-ships between money and nominaloutput.1 Understanding why themoney–income relationship canshift is critical to finding new waysof deriving information from money.

Why the Money–Nominal GDPRelationship Can Shift

A stable link between M2 andnominal GDP will hold as long aspeople handle their finances in thesame way.2 However, a marketeconomy will continuously createnew financial products and marketswill react to fundamental changes inthe tastes of households (Table 1 ).

Since the early 1980s, the attrac-tiveness to households of owningnon-M2 assets has increased be-cause of two types of technologicalchange: lower costs of transferringfunds from nonmonetary assets totransactions deposits (from bondmutual funds to money marketfunds, for instance) and greater use

“What growth in

conventionally measured

money means for inflation

will continue to change.”

B ecause inflation can quicklydisrupt an economy, central

banks have tried to develop poli-cies to keep inflation in check. Oneapproach assumes that there is astable relationship between eco-nomic activity and the measuredmoney supply. Recently, this rela-tionship has been changing becausepeople have been changing howthey handle their finances and howthey pay for goods and services. Asa result, what the measured moneysupply means, in terms of what itreveals about economic activity,has also changed.

Does M2 Still Measure Up?

Money and economic activity arelinked by the famous equation ofexchange:

money × money’s velocity

= the price level × real GDP,

or

M × V = P × Y.

In other words, changing hands Vtimes during a year, the money stock,M, facilitates the transaction of Ygoods, which each cost P dollars.Converting this equation into growth

inflationmoney supply growth

velocity growth

real output growth

= + −

nominal GDP

growth

money supply growth

velocity growth,

= +

6

of financial services from nonassetproducts (such as credit cards).Nonmonetary assets are any assetsnot included in the definition of themonetary aggregates, while nonassetproducts are instruments or waysof conducting transactions that donot directly and immediately involveholding an asset (for example, usinga credit card to pay for something)until final settlement is made. Asthe cost of shifting between non-M2assets and checkable deposits falls,the incentive to hold checkingdeposits to avoid transfer costsdeclines. Since households balancethe transfer cost savings from hold-ing money against the higher yieldson alternative assets, lower transfercosts have induced lower moneyholdings. For example, over the past10 years, the costs of shifting froma bond mutual fund to a checkablemoney market fund have fallen astransfer fees have fallen and astransfers have become easier. As aresult, when longer term interestrates (on bond funds) are high rela-tive to short-term rates (on moneymarket funds), people are morelikely to hold bond funds todaythan 10 years ago when transfersinvolved higher fees and greaterheadaches.

Thanks to improvements infinancial products, households and

firms can now better coordinatecash inflow with cash outflow. As aresult, they can reduce check usageby consolidating many purchasesinto fewer check payments. Theyalso have less need to hold checkingbalances for unexpected expenses.

Aside from technological changes,a rise in households’ awareness ofassets outside of M2 and theirtolerance for risk can lead to unusualweakness in M2. For example, ifhouseholds needed less extra returnon stocks to compensate them forthe extra investment risk, then at agiven gap between the yields onM2 and stocks, they will hold lessM2 and more stocks.

Technology and New Products

Lower asset transfer costs. Thecosts of shifting between non-M2and checkable M2 assets havefallen in several ways. First, load(commission) fees on mutual fundshave fallen sharply over the pasttwo decades.3 Furthermore, manymutual funds now also allow agreater number of free transfersamong funds in asset managementaccounts. These accounts offer ahost of investments, including bondsand equities, and allow no-costshifts among investments withinmutual fund families that typically

include a checkable money marketfund. So, a person who unexpect-edly gets hit with a big car repairbill can use the phone to shift fundsfrom an equity fund to a moneymarket fund (without incurring afee) and then write a money marketfund check. Furthermore, manybanks now offer mutual funds andallow customers to jointly managetheir mutual fund and depositbalances. Additionally, the FederalReserve has made it easier forpeople to buy Treasury securities,a change that, coupled with interestrates, encouraged people to takemoney out of M2 deposits and buyTreasury securities.4

More generally, the spread ofbetter information technology islowering transfer costs. In particu-lar, the rise of electronic banking(especially via personal computer)poses potentially large reductionsin the pecuniary and conveniencecosts of making such transfers.5

Unfortunately, continuous data onasset transfer costs over long periodsare lacking. Nevertheless, the limitedevidence implies that lower transfercosts have led people to reduce M2balances. In particular, lower transfercosts of using bond and equityfunds likely explains why most ofthe unusual weakness in M2 duringthe 1990s has been in small timedeposits (which compete with stocksand bonds) and money marketmutual funds (which were unusu-ally weak when relative yields onstocks and bonds yields were high).

Financial services from nonassets.In the 1970s and 1980s, techno-logical advances and high interestrates induced firms to avoid usingnon-interest-bearing demand de-posits to conduct transactions. Cashmanagement techniques, coupledwith the increased use of electronictransfers, allowed firms to moreeasily and cheaply tap nonmonetaryassets to meet cash shortfalls. Break-ing with the tradition of holding alot of non-interest-bearing demanddeposits, firms adopted cash man-agement techniques that enabledthem to better predict their cash

Table 1How Market Forces Can Cause Unusual Weakness in Money

Fundamental type of factor Examples

Technological innovations

Lower transfer costs Lower mutual fund commission (load) feesEasier purchase of Treasury securitiesElectronic bankingEasier banking and investing by phone

Financial services from nonassets More widespread credit cards and linesAutomatic teller machinesElectronic wires and transfers

Demographics, preferences and learning

Demographic shifts Rising population share of middle-agedpeople preparing for retirement

Preferences and financial sophistication Rising share of households with portablepensions due to IRA /Keogh laws andincreased job uncertaintyGreater tolerance of investment risk

7

needs. Also, firms increasingly usedwire transfers when they needed toshift funds. The result was a declinein demand deposits held by firms.

Financial innovations later spreadto households after improvementsin computer software made suchinnovations cost-effective for people.By providing liquidity and byenabling households to weathertemporary changes in asset prices(such as stock prices), credit cardsand credit lines likely induced manyhouseholds to hold less money andmore nonmoney assets.

For example, using 1983 data,Duca and Whitesell (1995) find thateach 10-percentage-point rise in theprobability of owning a credit cardlowers checking accounts by 9 per-cent and checkable money marketmutual funds and money marketdeposit accounts by 11 percent. Theimpact of credit cards on checkablebalances is likely larger today be-cause credit card ownership hasspread, credit cards are more widelyaccepted, credit card purchases aremore quickly processed, and con-sumers are now offered greater in-centives to use credit cards. Anotherimportant innovation is the spreadof automatic teller machines (ATMs).ATMs have reduced the need forpeople to carry extra cash by allow-ing them to easily withdraw cashfrom their checking or savingsaccounts.6

Evidence shows that becausepeople gained a greater choice inhow to pay for goods, the composi-tion of M2 had shifted away fromtransactions and toward nontrans-actions accounts. Coupled withlower transfer costs, greater use ofnonmoney ways of making paymentscould now be lowering M2, inaddition to altering its composition.

Are Demographics, PreferencesAnd Learning Playing a Role?

Greater tolerance of investmentrisk can stem from changes inemployment patterns, demograph-ics and in other factors that boostfinancial awareness.

Demographics. According to thelife-cycle theory of consumption,people borrow when they are youngbecause their income is below thatof later years, save in middle agewhen their income is highest andthen draw down their savings inretirement. An implication of thistheory is that savings rates and theshare of wealth invested in higherearning non-M2 assets should risein the peak earning years beforeretirement. By increasing the aver-age need to fund retirement, demo-graphic trends may be inducing anoverall shift toward risky assetswith higher expected long-termyields and away from lower earningM2 deposits. Alternatively, as peoplereach their peak earning years,their ratio of income to spendingfalls. As this ratio falls, so too will thepublic’s demand for low-transactionscost M2 deposits.

Consistent with these implica-tions, Duca and Whitesell (1995)find that small time and savingsdeposits are higher for older agegroups, after controlling for incomeand wealth. Furthermore, Morgan(1994) finds that the average shareof household assets held in stocksand bonds rises with the populationshare of 35- to 54-year-old people.

Changing preferences and learning.Two factors that could be depress-ing M2 holdings are households’increased awareness of investmentsoutside of M2 and an associatedrise in households’ willingness totolerate risk in the assets they con-trol. Aside from new technologyand financial products, increasedjob uncertainty and the liberaliza-tion of IRA/401K accounts haveinduced a shift toward portable(defined contribution) retirementplans that have given householdsa greater role in managing theirretirement assets. This shift, in turn,has induced households to incurlarge, one-time costs to learn moreabout bond and equity investmentsfor retirement. In addition, withmany mutual funds, people cancount their IRA/Keogh mutual fundbalances along with other mutual

fund holdings toward meeting theminimum balance requirements foropening asset management accounts.As a result, IRA and Keogh assetseffectively reduce the minimumbalance requirement on non-IRA/Keogh mutual fund assets. Consis-tent with this, both IRA/Keogh andnon-IRA/Keogh bond and equityfund assets rose in the mid-1980safter tax laws were eased and inthe early 1990s.7 Cross-section dataconfirm a big shift in householdportfolios toward bond and equityfunds and away from bank CDssince the late 1980s.8

Conclusion

The recent breakdown in thelink between nominal GDP andconventionally defined M2 reflectshow technological changes haveenabled households to hold lessmoney and more nonmonetaryassets. Such innovations have re-duced the costs of transferringfunds from other assets to checkingaccounts, or, as in the case of creditcards and lines, have reduced theneed to hold money that arises frommismatches of cash inflow andoutflow. Changes in tastes and theage composition of the U.S. popu-lation may also be heightening theextent to which people can substi-tute other financial assets for money.

The information revolution willlikely further reduce the benefitsfrom holding traditional forms ofmoney by fostering the spread ofnew electronic types of money,banking through personal computer,credit lines and financial manage-ment software. Together with theseadvances, a likely continuing shifttoward portable (defined contribu-tion) retirement plans and tax incen-tives will likely increase peoples’role in managing their retirementassets. These factors will likely leadpeople to further reduce theirholdings of conventionally defined“money” and increase their invest-ments in higher earning alternativeassets. As a result, what growth inconventionally measured money

8

Federal Reserve Bank of Dallas

Robert D. McTeer, Jr.President and Chief Executive Officer

Tony J. SalvaggioFirst Vice President and Chief Operating Officer

Harvey RosenblumSenior Vice President and Director of Research

W. Michael CoxVice President and Economic Advisor

Stephen P. A. BrownAssistant Vice President and Senior Economist

Research OfficersJohn Duca

Robert W. GilmerWilliam C. Gruben

Evan F. Koenig

EconomistsKenneth M. Emery

Beverly J. FoxDavid M. Gould

Joseph H. HaslagD’Ann M. Petersen

Keith R. PhillipsStephen D. Prowse

Fiona D. SigallaLori L. Taylor

Lucinda VargasMark A. WynneMine K. Yücel

Carlos E. Zarazaga

Research AssociatesProfessor Nathan S. Balke, Southern Methodist

University; Professor Thomas B. Fomby, SouthernMethodist University; Professor Gregory W. Huffman,

Southern Methodist University; Professor Finn E.Kydland, University of Texas at Austin;

Professor Roy J. Ruffin, University of Houston

Executive EditorHarvey Rosenblum

EditorsW. Michael Cox • Mine K. Yücel

Managing EditorRhonda Harris

Copy EditorMonica Reeves

Graphic DesignGene Autry • Laura J. Bell

The Southwest Economy is published six timesannually by the Federal Reserve Bank of Dallas. The viewsexpressed are those of the authors and should not beattributed to the Federal Reserve Bank of Dallas or theFederal Reserve System.

Articles may be reprinted on the condition that thesource is credited and a copy is provided to the ResearchDepartment of the Federal Reserve Bank of Dallas.

The Southwest Economy is available free ofcharge by writing the Public Affairs Department, FederalReserve Bank of Dallas, P.O. Box 655906, Dallas, TX75265-5906, or by telephoning (214) 922-5257.

means for inflation will continue tochange.

—John V. Duca

Notes

I thank the late Stephen Goldfeld and mymany colleagues throughout the FederalReserve System for sharing their insightson money with me over the years.

1 For examples, see Collins and Edwards(1994), Duca (1995a and 1994) andKoenig (1995).

2 For a more technical discussion, seeDuca’s (1995b) modified version ofMilbourne’s (1986) model of money.

3 For evidence, see Orphanides, Reid andSmall (1994).

4 See Feinman and Porter (1992).5 For more details, see Holland and Cor-

tese (1995) and Lewis (1995).6 Daniels and Murphy (1994a) find that a

100-percentage-point rise in the proba-bility of ATM use increased the velocityof currency (transactions/currency) by40 to 45 percent for transactions accountholders, while Daniels and Murphy(1994b) estimate that a 5-percent rise inthe proportion of ATM users would boostaverage transactions account balancesby 4.5 percent. Together, these studiesimply that ATMs induced householdsto shift from holding cash to holdingtransactions balances in the mid-1980s.

7 See Duca (1995a) for evidence.8 See Kennickell and Starr-McCluer (1994)

for cross-section evidence. These factorsare consistent with a study by Blanchard(1993), who found that the extra returnthat investors demand from equitiesover bonds has trended downwardsince the 1940s and abruptly fell in theearly 1980s.

References

Blanchard, Olivier J. (1993), “Movementsin the Equity Premium,” Brookings Paperson Economic Activity, no. 2: 75 –138.

Collins, Sean, and Cheryl L. Edwards(1994), “Redefining M2 to Include Bondand Equity Mutual Funds,” FederalReserve Bank of St. Louis Review,November/December, 7–30.

Daniels, Kenneth N., and Neil B. Murphy(1994a), “The Impact of TechnologicalChange on the Currency Behavior ofHouseholds: An Empirical Cross-SectionStudy,” Journal of Money, Credit, andBanking 26 (November): 867–74.

———, and ——— (1994b), “The Impactof Technological Change on TransactionsAccount Balances: An Empirical Cross-Section Study,” Journal of Financial

Services Research 17 ( January): 113 –19.

Duca, John V. (1995a), “Should BondFunds Be Included in M2?” Journal ofBanking and Finance 19 (April): 131–52.

——— (1995b), “Sources of Money Insta-bility,” Federal Reserve Bank of DallasEconomic Review, Fourth Quarter.

——— (1994), “Would the Addition ofBond or Equity Funds Make M2 a BetterIndicator of Nominal GDP?” FederalReserve Bank of Dallas Economic Review,Fourth Quarter, 1–14.

———, and William C. Whitesell (1995),“Credit Cards and Money Demand: ACross-Sectional Study,” Journal of Money,Credit, and Banking 27 (May): 604 –23.

Feinman, Joshua, and Richard D. Porter(1992), “The Continued Weakness in M2,”FEDS Working Paper no. 209, Board ofGovernors of the Federal Reserve System(Washington, September).

Holland, Kelley, and Amy Cortese (1995),“The Future of Money,” Business Week ,June 12, 66–78.

Kennickell, Arthur B., and Martha Starr-McCluer (1994), “Changes in FamilyFinances from 1989 to 1992: Evidencefrom the Survey of Consumer Finances,”Federal Reserve Bulletin, October, 861–82.

Koenig, Evan F. (1995), “Long-TermInterest Rates and the Recent Weaknessin M2,” manuscript, Federal Reserve Bankof Dallas, June.

Lewis, Peter H. (1995), “Chemical Aims toExpand Electronic Banking,” New YorkTimes, July 7, D5.

Milbourne, Ross (1986), “FinancialInnovation and the Demand for LiquidAssets,” Journal of Money, Credit, andBanking 18 (November): 506 –11.

Morgan, Donald P. (1994), “Will the Shiftto Stocks and Bonds by Households BeDestabilizing?” Federal Reserve Bank ofKansas City Economic Review, SecondQuarter, 31–44.

Orphanides, Athanasios, Brian Reid, andDavid H. Small (1994), “EmpiricalProperties of a Monetary Aggregate thatAdds Bond and Stock Funds to M2,”Federal Reserve Bank of St. Louis Review,November/December, 31–52.

Small, David H., and Richard D. Porter(1989), “Understanding the Behavior ofM2 and V2,” Federal Reserve Bulletin ,April, 244 –54.

9

Beyond the BorderA Look at the TopU.S. Trading Partners

T he U.S. economy grew 1.1 per-cent in the second quarter of

this year, down from 2.7 percent inthe first quarter. How much of arole has international trade playedin determining that growth, andwhat is it likely to contribute in thefuture? Over the past 14 years, U.S.trade (exports plus imports) as ashare of gross domestic product(GDP) increased from 8.8 percentto 17.8 percent. As the role of tradebecomes more significant in theU.S. economy, our trading partners’economies have a greater effecton the U.S. economy. This columnexamines what is happening nowin the economies of the largest U.S.trading partners and what 1996may hold.

Canada represents nearly 22percent of U.S. trade and is the topU.S. trading partner. Because theUnited States accounts for more than80 percent of Canadian merchan-dise exports, Canada’s economicfortunes are closely tied to those ofthe United States. In second-quarter1995, Canada’s real GDP fell 1 per-cent after a large drop in exports tothe United States. Capacity utiliza-tion declined in the second quarteras business inventories rose fromalready-high levels. Consumersalso spent less on big-ticket items.Probably the biggest source ofuncertainty has been Quebec’s questfor secession. Before the vote, thesecession referendum caused someuneasiness in financial markets, aseveryone tried to anticipate theoutcome. Despite current weakeconomic conditions, forecastersexpect real GDP to grow 3.1 percentin 1995 and 2.5 percent in 1996.

Japan’s economy appears to bein a state of uncertainty. Although

real GDP grew 3.1 percent in thesecond quarter after declining 0.1percent in the first quarter, othereconomic signals paint a differentpicture. Industrial production andcapacity utilization have been fallingand unemployment rising. Difficul-ties in the Japanese banking in-dustry are adding to the economy’swoes, and the government recentlyintroduced its sixth stimulus packagesince 1992 in hopes of jump-startingthe economy. Blue Chip forecasterspredict growth of 1 percent for1995 and 2.3 percent for 1996.

Mexico, the United States’ thirdlargest trading partner, is still re-covering from the December 1994peso devaluation. The country’sreal GDP fell 7.8 percent in the firstquarter of 1995, but only 3.1 per-cent in the second quarter, on aseasonally adjusted, quarterly changebasis. A bright spot is the drop ininterest rates on cetes, peso-denomi-nated debt issued by the Mexicangovernment. The interest rate on28-day cetes was 40.6 percent onOctober 19, down dramaticallyfrom the 80-percent high in April.

Mexico’s economic indicatorsshow some negatives, however.Industrial production continues tofall and is nearing the lowest rateof the 1990s, while unemploymentis rising to the highest rate of thedecade. Although Mexico recentlyrepaid $700 million of its debt to theUnited States, financial markets con-tinue to rain on Mexico’s parade.From September 4 to October 18,Mexico’s Bolsa stock exchange de-teriorated 10 percent and the pesofell 7 percent. For the year, theOrganization for Economic Coopera-tion and Development (OECD)expects Mexican real GDP to de-cline by 3 to 4 percent. Growthshould resume in 1996, however,at a rate of 2.5 percent.

Germany experienced 2.2-percentgrowth in the second quarter, down

from 2.9-percent growth in the firstquarter and 3-percent growth for1994. The unemployment rate re-mains high, while industrial pro-duction dropped significantly inAugust after steadily increasing inthe past four months. Competitivepressures, however, are helpingliberalize some parts of the economy.For example, laws preventing retailstores from selling past 6 p.m. arebeing challenged, and there is amovement to ease the tax burdenon businesses. The OECD expectsreal GDP to grow by 2.9 percent in1995 and 2.7 percent in 1996.

The United Kingdom experi-enced real GDP growth of 1.8 per-cent in the second quarter, downfrom 2.6 percent in the first quarter.Analysts attribute the slowdown toa sharp drop in exports. On a morepositive note, the unemploymentrate has been steadily decreasingover the past two years. Capacityutilization remains far above aver-age, and industry surveys havereported more than half of all firmsare working at full capacity. RealGDP is expected to grow 3 percentin 1995 and 2.6 percent in 1996.

So, what does this all mean forU.S. exports and growth? This year,U.S. exports to Japan, Germany andCanada have surpassed 1994 levels,while exports to Mexico and theUnited Kingdom have lost ground.The outlook for next year is some-what mixed as well. Mexico is ex-pected to start its recovery nextyear, although uncertainty is stillundermining the economy.

Canada, Germany and the UnitedKingdom should grow at a healthyrate in 1996. Forecasters expectJapan’s global trade surplus to con-tinue to decline throughout 1995and 1996. Given Japan’s weakness,however, its imports from the U.S.may be weak.

— David GouldMichelle Thomas

10

Regional Update

FURTHER INFORMATION ON THE DATAFor more information on employment data,

see “Reassessing Texas Employment Growth”(Southwest Economy, July/August 1993). Formore information on TIPI, see “The Texas Indus-trial Production Index” (Dallas Fed EconomicReview, November 1989). For more informationon the Texas Leading Index and its components,see “The Texas Index of Leading Indicators:A Revision and Further Evaluation” (Dallas FedEconomic Review, July 1990).

Online economic data and articles are avail-able on the Dallas Fed’s electronic bulletin board,Fed Flash (214) 922-5199 or (800) 333-1953.

REGIONAL ECONOMIC INDICATORSTexas Employment Total Nonfarm Employment

Texas PrivateLeading TIPI Construc- Manufac- Govern- service- NewIndex total Mining tion turing ment producing Texas Louisiana Mexico

9/95 — — 154.4 412.3 1,035.0 1,463.1 5,042.9 8,107.7 1,796.0 694.68/95 113.2 119.2 153.8 407.6 1,034.4 1,466.4 5,037.7 8,099.9 1,791.2 691.17/95 112.9 119.6 154.1 405.2 1,031.9 1,458.5 5,020.0 8,069.7 1,786.8 690.36/95 112.6 119.2 155.3 404.2 1,032.3 1,451.9 5,002.9 8,046.6 1,787.6 686.65/95 112.4 119.1 155.7 402.7 1,031.3 1,448.1 4,980.7 8,018.5 1,786.5 686.54/95 111.2 118.8 156.1 403.7 1,033.4 1,444.6 4,975.2 8,013.0 1,783.2 685.83/95 110.1 118.7 156.7 406.3 1,032.2 1,441.0 4,941.8 7,978.0 1,784.4 685.32/95 111.1 119.1 157.3 405.7 1,029.7 1,438.3 4,930.7 7,961.7 1,782.4 684.31/95 110.4 118.9 157.6 402.0 1,026.5 1,433.3 4,914.0 7,933.4 1,781.8 681.5

12/94 111.4 118.3 158.9 398.8 1,025.0 1,433.2 4,916.9 7,932.8 1,774.5 675.311/94 111.9 118.2 160.1 393.0 1,021.7 1,426.8 4,893.0 7,894.6 1,764.0 674.210/94 112.0 118.5 160.6 388.2 1,018.8 1,424.5 4,865.5 7,857.6 1,755.1 669.0

Total Nonfarm Employment

Index, January 1991 = 100

Texas Industrial Production Index

Texas Leading Index and Nonfarm Employment

Index, January 1991 = 100

United States

New Mexico

LouisianaTexas

1995199419931992199195

100

105

110

115

120Total

Utilities

Mining

Manufacturing

85

90

95

100

105

110

115

19951994199319921991

Net Contributions of Components to Change in Leading Index, June–August 1995

Percent

Average weekly hoursHelp-wanted indexTexas stock index

New unemployment claimsReal retail sales

Well permitsReal oil price

U.S. leading indexTexas value of the dollar

–.10.03

–.06

.03

.06

.87

–.22

–.12

–.4 1

Thousands of persons Index, January 1981 = 100

Leading indexEmployment

’95’93’91’89’87’85’83’8185

90

95

100

105

110

115

5,500

6,000

6,500

7,000

7,500

8,000

8,500

–.2 0 .2 .4 .6 .8

.27

The Eleventh District economy isgrowing at a moderate pace. Employ-ment in Louisiana accelerated slightlyfrom July to September, following weakgrowth since January. New Mexicoemployment growth picked up in thethird quarter, after weakness in thesecond quarter. Texas nonfarm employ-ment growth slowed in September,following strong growth in the priorthree months. The most recent BeigeBook survey of District business con-ditions also suggests slightly weakergrowth in Texas.

Manufacturing remains one of theweakest sectors of the District economy.Texas manufacturing production de-clined in August after a mild pickup in

June and July. Manufacturing employ-ment in the District states increasedonly slightly in the third quarter. Weak-ness in the national economy this yearand a decline in exports, particularly toMexico, have led to weakness in themanufacturing sector. In the secondquarter, a 10-percent decline in exportsto Mexico caused total Texas exportsto decline 2.2 percent. Construction-related manufacturing industries havealso been weak, although a recentpickup in residential building shouldresult in increased orders for theseindustries over the next six months.

A recent pickup in new home build-ing has boosted District constructionemployment, which surged in August

and September. Despite declines in thefirst quarter, construction employmentlevels remain higher than the stronglevels posted a year ago. District busi-ness contacts report that lower mort-gage rates and declines in home pricesin many markets have spurred theturnaround.

The Texas leading index increased inAugust for the fifth straight month. Theindex has recovered from sharp declinesin the first quarter that were mainly dueto the peso devaluation. In May, theindex surpassed its previous peak andhas since experienced steady growth.Five of the nine indicators increased inthe three months ending in August, ledby solid gains in stock prices and retailsales. Recent movements in the indexsuggest that moderate expansion willcontinue over the next six months.

—Keith R. Phillips

11

BULK RATE

U.S. POSTAGE

P A I DDALLAS, TEXAS

PERMIT NO. 151

FEDERAL RESERVE BANK OF DALLASP.O. BOX 655906

DALLAS, TEXAS 75265–5906

The Challenge of NAFTAAn international economic forum evaluating NAFTA’s impact almost two years after the treaty’s passage. Panelists will examine international trade

and investment in the NAFTA countries, financial interdependencies, macro-industrial and micro-industrial development, and retail trade.

WHEN: December 3, 6:30 p.m.–8:30 p.m. (reception) and December 4, 9:00 a.m.–6:30 p.m. (program)PLACE: Camino Real Paso del Norte Hotel, 101 S. El Paso St., El Paso, Texas 79901

REGISTRATION DEADLINE AND FEE: November 24, 1995; $90 per personFOR MORE INFORMATION: Leigh Bloss, El Paso Branch–Federal Reserve Bank of Dallas, (915) 521-8235, Fax (915) 521-8284

PROGRAM HIGHLIGHTS• Addresses by Raul Hinojosa-Ojeda, UCLA; Dallas Fed President Robert D. McTeer, Jr.; and Ariel Buira, subgovernor, Banco de México.• NAFTA and International Trade and Investment with Oscar Vera Ferrer, Centro de Estudios Económicos del Sector Privado; Alan Rugman,

University of Toronto; and Dallas Fed economists Harvey Rosenblum, Lucinda Vargas and William C. Gruben.• NAFTA and Financial Interdependence with William C. Gruben, Alfredo Phillips Olmedo, North American Development Bank; Sylvia Maxfield,

Yale University; Catherine Mansell Carstens, Instituto Tecnológico Autónomo de México.• NAFTA and Macro-Industrial Development with Timothy P. Roth, UT El Paso; Chuck Cunningham, Delphi Packard Electric Systems — General

Motors; Alberto Sandoval, Internacional de Cerámica; Carlos de Orduña, Sanyo North America Corp.; José Reyes Ferriz, Reyes, Estrada y Fernández• NAFTA and Micro-Industrial Development with Lucinda Vargas; Hector Moreira, Instituto Tecnológico y de Estudios Superiores de Monterrey;

Luis Ortega, Kokopeli; Gabriela Quirarte, D.M. Distillery; Ed L. Romero, Advanced Sciences.• NAFTA and Retail Trade with Frank Hoy, UT El Paso; Nora Yu, Cámara Nacional de Comercio de Cd. Juárez; and Michael Patrick, Texas A&M

International University.

Sponsored by the El Paso Branch of the Federal Reserve Bank of Dallas, in conjunction with the University of Texas at El Paso, College of Business;New Mexico State University, Border Research Institute; Universidad Autónoma de Cd. Juárez, Instituto de Ciencias Sociales y Administración;

and Universidad Autónoma de Chihuahua, Departamento de Vinculación y Transferencia de Tecnología.