Why We Undervalue Tomorrow: Time Preference From Austrian Economics to the Psychology Lab

The alarm goes off and the snooze button wins. The hard conversation moves to next week, again. None of that makes us bad people; it makes us human.

Economists have a name for this; psychologists have a few more. The name is time preference: how much we value now versus later.

I’m a therapist and an author, and I love this idea because it lives in two worlds at once. Carl Menger, Eugen von Böhm-Bawerk and Ludwig von Mises wrote about it as economics, and two authors I keep coming back to, Hans-Hermann Hoppe and Saifedean Ammous, carried it forward. Psychology studies it in the lab under other names: delay of gratification, delay discounting and self-regulation. I want to walk through both honestly, including the parts that don’t hold up.

What the economists saw

Menger put it plainly in his Principles of Economics, first published in German in 1871: “All experience teaches that a present enjoyment or one in the near future usually appears more important to men than one of equal intensity at a more remote time in the future.” He argued that our most anxious care goes to the present and the near future, and that this concern fades as the horizon gets longer. In his words, “This phenomenon is not accidental but deeply imbedded in human nature.”

Böhm-Bawerk picked up the thread in The Positive Theory of Capital, published in German in 1888 and 1889. His line is one of the clearest I know: “We systematically underestimate future wants, and the goods which are to satisfy them.”

What I love about Böhm-Bawerk is that he tried to explain why. He named three grounds. First, “the incompleteness of the imaginations we form to ourselves of our future wants.” Second, “a defect in will.” Third, “the shortness and uncertainty of life.” He also did something humble; he wrote that psychologists seemed to him “more competent to decide” these questions than economists.

Mises, in Human Action (1949), made time preference foundational. “Present goods are more valuable than future goods,” he wrote. “Time preference is a categorial requisite of human action.” He also wrote one of the most human sentences in economics: “Time for man is not a homogeneous substance of which only length counts.” A minute from now and a year from now don’t feel the same; in his words, they “appear in different perspective” depending on how near they are to the moment we decide.

Here’s the part that gives me hope. Mises noted that time preference isn’t unique to humans; animals show it too. But in people it “is not inexorable”; how far ahead we provide is “the result of a process of valuation.” In other words, it’s something we weigh, not only something that happens to us.

Hoppe, in Democracy: The God That Failed, describes time preference as “a universal preference for earlier over later goods, and for more over less durable ones.” I use Hoppe here only for his time preference economics. He argues that each person sets their own rate according to their subjective evaluations, and that its level and its changes over a lifetime “depend on personal psychological factors.” He also argues that once people can save at all, time preference tends to fall, a self-reinforcing trend he calls a “process of civilization,” and that as planning horizons widen, so does the interest in knowledge. That’s his argument; none of the sources I’m drawing on test it.

Ammous calls time preference “a pivotal concept” in The Bitcoin Standard. His thesis is that money which holds its value across time “gives people a bigger incentive to think of their future, and lowers their time preference.” I’m sharing that as his argument about money, not as a proven finding, and not as advice of any kind.

The honest twist

You might expect psychology to show up and prove the Austrians right. Mises would push back. He criticized Böhm-Bawerk on exactly this point: “His demonstration of the universal validity of time preference is inadequate because it is based on psychological considerations. However, psychology can never demonstrate the validity of a praxeological theorem.”

So I treat these as parallel lenses. The economics describes a logical feature of action. The psychology measures tendencies in real people, with real noise. They rhyme; one does not prove the other.

What psychology measures

Shamosh and Gray describe self-control as a tendency or ability to favor “distal, more important goals over proximal, less important ones.” An important indicator of it is delay discounting: “the tendency to prefer smaller, sooner rewards to larger, later ones.” Psychologists even use the phrase “time preference” for one family of these measures, such as how often someone picks the immediate reward. Same words; not quite the same meaning.

The marshmallow story, updated

Most of us have heard of the marshmallow test. In 1989, Mischel, Shoda and Rodriguez reported in Science that 4-year-olds who waited longer in certain lab situations grew into more cognitively and socially competent adolescents, with higher school performance and better coping with frustration and stress.

That story is now contested. In 2018, Watts, Duncan and Quan ran a conceptual replication with 918 children, focusing on 552 children of mothers without a college degree. One extra minute of waiting at age 4 predicted about a tenth of a standard deviation higher achievement at 15. That was half the size of the original reports, and it shrank by two thirds once they controlled for family background, early cognitive ability and the home environment. Most of the effect came from being able to wait at least 20 seconds; links to behavior at 15 were much smaller and rarely statistically significant. The authors were careful; they did not claim causal effects.

In 2020, Doebel, Michaelson and Munakata argued that many of those controls captured processes that support waiting in the first place, so a weaker link was “not surprising.” Watts and Duncan replied the same year. My fair summary: waiting seems to matter somewhat, the effect looks smaller than the legend, and researchers still disagree about how to read it.

Kidd, Palmeri and Aslin (2013) ran the marshmallow task with an experimenter who was either reliable or unreliable. Children in the reliable condition waited significantly longer. The authors suggest the wait times reflected “reasoned beliefs about whether waiting would ultimately pay off.” It was a small study of 28 children; still, it rhymes with Böhm-Bawerk’s third ground. Children, at least, wait longer when the future looks trustworthy.

Patience and intelligence: a modest link

People tend to overstate this part, so let me be careful. A 2008 meta-analysis by Shamosh and Gray pooled 24 studies with 26 effect sizes. Higher intelligence was associated with lower delay discounting, with a weighted mean correlation of r = −0.23, and they found no evidence of publication bias. The authors call it “small-to-moderate.” Squared, that means the two share roughly 5% of their variance.

A separate 2008 study of 103 adults, by Shamosh and colleagues, found that delay discounting was negatively correlated with general intelligence and working memory, partly through activity in the left anterior prefrontal cortex.

Correlation only. None of this shows that intelligence causes patience or that patience causes intelligence; the direction is unknown. It says nothing about anyone’s worth.

The Dunedin study adds another angle. Moffitt and colleagues followed 1,000 children to age 32 and found that childhood self-control predicted physical health, substance dependence, personal finances and criminal offending along a gradient, and that those effects could be disentangled from intelligence and social class. It’s observational too; still, it’s a reminder that self-control isn’t just intelligence in disguise.

Böhm-Bawerk’s first ground, tested

Remember Böhm-Bawerk’s first ground, our incomplete imagination of future wants? He didn’t run experiments; researchers now test something very close to it, called episodic future thinking: vividly imagining a specific future event.

In a 2010 brain imaging study, Peters and Büchel found that episodic future thinking reduced the rate of delay discounting, and a second experiment replicated the behavioral effect. A 2022 meta-analysis by Rösch, Stramaccia and Benoit, across 174 effect sizes from 48 articles, found a medium effect (g = .44); it was stronger when the imagined events were positive, more vivid and related to the choice. A second 2022 meta-analysis by Ye and colleagues found a similar effect (g = 0.52), larger for positive imagery (g = 0.64) and essentially zero for neutral or negative imagery (g = −0.03). Those two reviews likely draw on many of the same studies, so I count them as one strong signal rather than two. A 2026 meta-analysis by You and colleagues pooled 16 randomized trials with 881 people in real-world settings and found that it reduced delay discounting too (SMD = −0.35); the authors say long-term effects still need study.

The wider view comes from a 2018 review by Rung and Madden. Many methods reduce discounting; framing and priming mostly shift one-off choices. In their words, “The largest and longest-lasting effects were produced by learning-based manipulations.” The summary I’m drawing on doesn’t spell out which learning methods those are, so I won’t pretend it does.

What I take from all this

These are educational ideas, small and practical. Expect lab-sized effects, not a personality transplant.

1. Picture one specific moment

   Pick a positive, vivid moment tied to the choice in front of you. Not “being healthier someday”; the actual morning you want to wake up to. Of everything here, this has the most research behind it.

2. Move your attention

   Rodriguez, Mischel and Shoda (1989) found that children who distracted themselves from the tempting parts of the situation waited longer, even with age and intelligence controlled. That study was in older children with social adjustment problems, not adults.

3. Practice; don’t just decide

   Learning-based methods showed the largest and longest-lasting effects. My reading is that patience is something we build through repetition rather than one big decision. That’s my interpretation, not a specific protocol from the research.

4. Keep promises to your future self

   This one is my view, not a research finding. If kids wait longer when the adult in the room is reliable, I suspect we wait longer when we’ve learned that we ourselves are reliable. Small promises, kept, may make the future feel worth waiting for. I haven’t seen it tested; treat it as a hypothesis.

Go deeper, not bigger

Menger saw it, Böhm-Bawerk tried to explain it, Mises made it a category of action, and psychology keeps measuring it. My take: lower time preference isn’t a trait you either have or lack; it’s a practice. Start with one vivid moment this week. You got this.

Sources

1. Carl Menger, Principles of Economics (1871; trans. Dingwall & Hoselitz, 1950; Ludwig von Mises Institute reprint, 2007), pp. 153 to 154. cdn.mises.org
2. Eugen von Böhm-Bawerk, The Positive Theory of Capital (1888/89; trans. William Smart, 1891), Book V ch. III, pp. 253 to 255. cdn.mises.org
3. Ludwig von Mises, Human Action: A Treatise on Economics, Scholar’s Edition (1949; Ludwig von Mises Institute, 1998), ch. XVIII “Action in the Passing of Time”, sec. 2 “Time Preference as an Essential Requisite of Action”, pp. 480, 481 and 485 (including n. 4). cdn.mises.org; wording also checked against the mises.org online edition (4th ed., different page numbers): mises.org
4. Hans-Hermann Hoppe, Democracy: The God That Failed (2001; Routledge 2017), ch. 1, pp. 1 to 7. pageplace.de preview
5. Saifedean Ammous, The Bitcoin Standard (Wiley 2018; 2021 printing), ch. 1 p. 7 and ch. 5 p. 73. Official sample: shopify CDN; ch. 5 preview: oreilly.com
6. Shamosh & Gray (2008). Delay discounting and intelligence: A meta-analysis. Intelligence, 36(4), 289 to 305. doi.org/10.1016/j.intell.2007.09.004
7. Shamosh et al. (2008). Psychological Science, 19(9), 904 to 911. doi.org/10.1111/j.1467-9280.2008.02175.x
8. Mischel, Shoda & Rodriguez (1989). Science, 244(4907), 933 to 938. doi.org/10.1126/science.2658056
9. Watts, Duncan & Quan (2018). Psychological Science, 29(7), 1159 to 1177. PMC6050075
10. Doebel, Michaelson & Munakata (2020). Psychological Science, 31(1), 97 to 99. PMC10152116; Watts & Duncan reply (2020), Psychological Science, 31(1), pp. 105 to 108: doi.org/10.1177/0956797619893606
11. Kidd, Palmeri & Aslin (2013). Cognition, 126(1), 109 to 114. PMC3730121
12. Moffitt et al. (2011). PNAS, 108(7), 2693 to 2698. PMC3041102
13. Rodriguez, Mischel & Shoda (1989). Journal of Personality and Social Psychology, 57(2), 358 to 367. PMID 2760808
14. Peters & Büchel (2010). Neuron, 66(1), 138 to 148. PMID 20399735
15. Rösch, Stramaccia & Benoit (2022). Journal of Experimental Psychology: General, 151(7), 1606 to 1635. PMID 34843367
16. Ye et al. (2022). Quarterly Journal of Experimental Psychology, 75(10), 1876 to 1891. PMID 34841982
17. You et al. (2026). International Journal of Nursing Studies, 182, 105619. PMID 42341710
18. Rung & Madden (2018). Journal of Experimental Psychology: General, 147(9), 1349 to 1381. PMC6112163