Is College Still Worth the Time and Expense?

Though U.S. residents increasingly question whether a college education is worth the time and expense,  it likely remains true that for any number of “white collar” or “office” jobs, a college degree does matter as a signaling device and sorting procedure for employers. 

In other words, a college degree acts as a signal to employers that a person possesses "unobservable" qualities like perseverance, conscientiousness, and the ability to delay gratification. That is why, in many cases, the actual degree subject matter does not matter. 

It is the personal habits employers believe a college graduate possesses that matter. 

Study	Key Findings	How it Connects to "Signaling"
Spence (1973), "Job Market Signaling"	Seminal work in economics that introduced the signaling theory as a concept to explain labor market outcomes. It posits that education is a signal of a worker's innate ability to employers.	The model shows that in a world of imperfect information, high-ability workers invest in education to signal their productivity to employers and distinguish themselves from low-ability workers. The degree itself is the signal, not the knowledge it contains.
Tyler, Murnane, and Willett (2000)	Explored the "sheepskin effect," which is the wage premium associated with completing a degree, compared to just having a similar number of years of schooling.	Found that there is a significant jump in earnings for those who complete their degree (the "sheepskin") versus those who drop out with similar total years of schooling. This suggests that the signal of completion is more valuable than the simple accumulation of credits or knowledge.
Bingley, et al. (2015)	Using a panel of Danish twins, researchers found that the wage return to education decreases when controlling for unobserved factors like "ability," suggesting these factors are highly correlated with schooling.	By comparing identical twins with different educational attainments, the study attempts to isolate the effect of education from genetic and shared environmental factors (which are strong proxies for unobservable skills). The finding that the return to schooling is smaller when accounting for these factors supports the idea that the degree is partly a signal for pre-existing ability.
Arteaga (2018)	Found that a reform at a Colombian university that reduced the "human capital" content of a degree (i.e., less specific knowledge) had a significant negative impact on the graduates' earnings, especially for those from lower-reputation institutions.	This study shows a blend of both human capital and signaling theories. It suggests that while the degree serves as a general signal of quality, the specific skills gained (human capital) are also important. The "signaling" effect is especially crucial for graduates of less prestigious schools, who rely more on their degree to gain access to higher-paying firms and better job matches.

That noted, lots of well-paying jobs do not require such signaling mechanisms.

If You Want to Make the World a Better Place, Take a Look at Yourself and Make the Change

Michael Jackson: "I'm starting with the man in the mirror: I'm asking him to change his ways. And no message could have been any clearer:

If you wanna make the world a better place
Take a look at yourself and then make a change.

We have met the enemy. It is us. Each of us. 

Even if AI Creates Abundance, Does it End Zero-Sum Economics?

AI's ability to reduce the cost of intelligence nearly to zero might be thought  to fundamentally challenge zero-sum economic principles (what one entity gains another loses) in many domains, including content. 

But it will probably be a complex matter, as has been the case for the way internet-fueled abundance affected the economics of content markets where scarcity has been the rule. 

Consider the impact on content. The internet fundamentally transformed content economics by collapsing the barriers that once made content creation and distribution expensive and exclusive. 

Traditional media businesses built their models around scarcity: limited time, space and distribution real estate. The internet changed all that, enabling much lower cost content creation and distribution. 

The key point is that scarcity-based content business models became vulnerable when the internet removed the underlying scarcity. 

But new forms of value creation emerged around curation, community, convenience, and trust. So sources of value shifted, in many cases. But abundance has not actually eroded zero-sum economics, just reshaped it. 

Consider attention, which is genuinely finite and zero-sum. Each minute spent watching YouTube is a minute not spent reading newspapers or watching television. Traditional media's loss of audience attention directly translates to digital platforms' gain. 

This explains why established media companies have struggled so intensely with the digital transition. They're fighting for shares of a fixed attention economy.

However, other aspects of digital markets can be positive-sum. The internet enabled entirely new forms of value creation that didn't exist before. Wikipedia didn't just redistribute encyclopedia sales to volunteers - it created an entirely new model of collaborative knowledge. 

It isn’t yet clear where AI might actually break the zero-sum model, as AI doesn't fully escape fixed-resource constraints. 

Zero-sum games will persist in many areas, since not all resources become abundant, and might well persist for other reasons, such as a shift in scarcity value to attention. 

In many cases, physical limits will remain, for example. There are only going to be so many National Football League regular-season games. There will only be one SuperBowl. 

So paradoxically, perhaps, the era of content abundance might not eliminate zero-sum economics as much as one would think. 

Where all major contestants in an industry (think law, marketing, manufacturing, transportation or finance, for example) adopt AI, there is no net economic advantage. Rivalry continues, and no firm gains a sustainable competitive advantage. 

Content abundance does not mean an end to strong zero-sum pressures in television, radio, movies or music. Popularity still will be quite unevenly distributed. 

In the professional content businesses (music, film, video, books, magazines, newspapers, radio) the issue is “what is scarce?” Traditionally, we might argue that content creation is expensive, and therefore scarce. 

In the era of digital content, it is human or audience attention that is finite, and therefore scarce, whereas the supply of content approaches infinity. That might continue to be true even if AI enables new content formats that gain favor with consumers. They will simply consume less content in other formats. 

Traditionally, content markets have had zero-sum elements because producing high-quality content required scarce resources like time, skill, and capital. 

AI, as did digital media and the internet before it, disrupts scarcity value by making content generation cheap, fast, and scalable.

The “oversupply” of content effectively collapses the scarcity that once gave content its intrinsic value. As the marginal cost of creating content approaches zero, the content itself becomes a commodity.

So a creator's success is no longer determined by their ability to create, but by their ability to get noticed i

The value proposition shifts from "what can I create?" to "what can I get people to pay attention to?"

That is the attraction of live sports content, for example. 

The point is that AI support for abundant content creation does not necessarily mean an end to zero-sum economic effects in media industries. Consumers have only so much time available to consume content.

Who Knew GPS Spoofing was a Thing for Uber and Lyft?

Who even knew GPS spoofing was a thing? Uber and Lyft, for starters, as their business models hinge on accurate location information to connect riders and drivers or food and product buyers with sellers. But the ride sharing companies know some drivers manipulate their location data for personal financial gain, to the detriment of other honest drivers, passengers and the ride sharing companies. 

It is difficult to estimate how extensive such spoofing might be, although in likely rare cases, the financial damage can run to scores of millions of dollars.  

But GPS feeds can be disrupted for any number of reasons, such as a vehicle entering a tunnel. But there are other ways for GPS location data to be inaccurate in urban areas with lots of tall buildings or tree cover. The algorithms ride sharing companies use to detect such anomalies mean some honest drivers will occasionally be de-platformed by mistake. 

Some Uber and Lyft drivers manipulate their GPS location through "GPS spoofing" to increase their earnings, as the apps use location to determine trip requests, fares, and bonuses, which gives drivers an incentive to try and "game" the system. 

In some cases, a coordinated group of drivers in a specific area, such as an airport queue, will all go offline simultaneously, causing a sudden drop in the number of available drivers, triggering a "surge" in fares for that location.

In a similar way, cheaters can use GPS spoofing to place themselves virtually in the airport queue, even when they are miles away.

In other cases, drivers might use GPS spoofing or multiple accounts to exploit incentive programs. A driver could create multiple fake passenger accounts on the same device and use GPS spoofers to create fraudulent trips to meet quotas and collect bonuses.

There are a range of ways to do such things. Drivers can use "fake GPS" applications, which does not require technology skill. Others might use “rooted devices” with access to operating systems, to use more sophisticated spoofing methods that are harder for apps to detect. 

Some techniques use multiple phones. Scammers might use one phone with a spoofed location to get into an airport queue virtually, then use a second phone to accept and complete trips in their actual location.

And while the potential revenue might not always be worth the investment, dishonest drivers can GPS signal generators that transmit fake GPS signals.

As always is the case, virtually any technology can be used for good or evil; honest purposes or dishonest.

For Every Public Purpose There are Corresponding Private Interests: AI Data Centers are No Exception

Many discussions of large artificial intelligence data center investments focus on the expected benefits of job creation. But there also are costs, which might include the lost tax revenue for states which offer tax incentives to data center operators; higher consumer power bills; higher water consumption; stranded utility assets and even far fewer new jobs than anticipated. 

In that regard, large AI data centers remind me of large sports stadiums: the argument is that government subsidies will promote economic growth. In fact, such investments mostly shift consumer spending from one category to another, with possible net zero gains. 

Study / Source	Main Finding on Net Economic Benefits
Baade and Dye (1990), Growth & Change	No consistent evidence that stadiums boost metropolitan income or employment.
Noll & Zimbalist (eds.), Sports, Jobs & Taxes (1997)	Broad survey finds subsidies rarely generate promised jobs or growth; most benefits overstated.
Coates & Humphreys (2008 review)	Majority of peer-reviewed studies show little to no positive impact on city-wide income/employment.
Baade and Matheson (2000, 2004) – Super Bowl studies	Event impacts overstated; spending largely shifted from other activities, little net new growth.
Matheson (2018, J. Policy Analysis & Management)	Citywide gains minimal; limited neighborhood or intangible benefits possible, but not enough to justify heavy subsidies.
Mercatus Center  (2015 working paper)	Finds no evidence subsidies increase growth; in some cases correlate with lower growth.
Bradbury, Coates, Humphreys (2022/2023 survey)	Comprehensive literature survey: consensus remains that stadium subsidies almost never deliver large net benefits.
Policy briefs (Journalists’ Resource, CBCNY)	Ex-ante booster studies overstate benefits; opportunity costs of public spending usually higher than stadium returns.

That is not to deny the need for large AI data centers, simply to point out that the local economic benefits might not be as often touted.

Study/Report	Year	Promised/Expected Jobs	Actual/Direct Jobs	Key Findings
Upwind Industry Research	2025	N/A (hype around economic catalysts)	1,688 (construction, temporary); 157 (permanent operations)	Construction jobs are short-term and high-volume, but permanent roles are minimal relative to facility size (~$243M economic add during build vs. $32.5M annually ongoing); underscores automation's role in limiting ops staffing.
University of Michigan / STPP Report ("What Happens When Data Centers Come to Town?")	2025	High-paying jobs as primary benefit (per subsidy pitches)	Few permanent positions (not quantified)	Jobs generated are disproportionately low compared to subsidies (e.g., billions in tax breaks); recommends redirecting funds to education or renewables for better ROI; no broad economic uplift observed.
Wall Street Journal (on Stargate AI Project)	2025	Employment "bonanza" (tech/political hype)	~100 full-time	Lowest jobs per square foot of any major facility (1M sq ft campus); contrasts with 500 jobs at a nearby 286k sq ft cheese plant—exposes AI boom as labor-light post-construction.
Good Jobs First (analysis of subsidies) - Google Indiana	2025	200 jobs (governor's announcement)	30 (local tax abatement minimum; no state mandate)	Massive gap between promises and enforceable commitments; subsidies flow without job guarantees, prioritizing corporate gains over local employment.
Good Jobs First (analysis of subsidies) - Amazon Indiana	2025	>1,000 jobs	400 direct (plus 600 subcontractors)	Subcontractor roles often lack benefits/wage parity; highlights "illusory" totals and long timelines (e.g., spread over years), diluting immediate impact.
Oxford Economics (for Google Data Centers)	2018	N/A (focus on multipliers)	1,900 direct (national); ~700-1,700 total per state	Positive multiplier (3.3-4.6x, national 5.9x) boosts indirect jobs to 11,000 total, but direct ops roles remain low (~100-200 per campus); early study predates AI surge but shows pattern of limited core employment.
CBRE Economic Ripple Effects	2024	N/A	3.5M direct-related (up 20% from 2017)	Strong growth but concentrated (e.g., 490 ops + 1,500 construction in NE, 2022); 7.4x multiplier inflates totals, yet critics note it masks low direct density in AI era.

Then there are the other issues such as higher consumer electricity bills and impact on water usage. 

McKinsey estimates suggest that by 2030, data centers globally will  require $6.7 trillion in investment for compute operations, of which $5.2 trillion in capital expenditures will support artificial intelligence operations. 

If correct, that represents nearly $7 trillion in capital outlays by 2030, and would be an increase of 3.5 times the capacity of data centers from 2025 to 2030 alone. 

source: McKinsey 

source: McKinsey

Data center power needs in the United States alone are expected to add about 460 terawatt-hours of demand from 2023 to 2030, three times the current level of consumption, McKinsey estimates. At the same time, data center water demand could rise about 170 percent by 2030, according to analysts at WestWater Research. 

Those forecasts might be wrong on the high side, but even so, much to all of that capacity will mostly have to be built, somewhere.

And the point is that the benefits and costs will accrue to different participants in the information technology value chain, in different quantities. There will probably be less benefit for local economies, taxpayers and electricity and water ratepayers than often is assumed.