The common perception behind eating on a budget is that you’ll have to resort to cheap and unhealthy food.
If you go on any nutrition or frugality forum you’ll see this question come up often, with mixed answers of:
So which is it?
What I Set Out to Do
Finding out the answer to that question is my primary motivation.
In order to find as complete an answer as possible, I compiled a list of foods with their nutrient density and their typical cost, which led to some interesting findings that I’ll now present.
How it was Done
A while back I saw the presentation below done by Mathieu Lalonde, where he looked at the nutrient density of several thousand foods. My intention was to build off of the work he did here:
By the time I had gone through all of the foods in the nutrition spreadsheet, I was left with just over 1000 different foods complete with vitamin/mineral information and typical pricing. I’ll include a link to the spreadsheet at the end of this post for download.
The main ‘score’ I calculated was as follows:
This presented a bit of a problem when there were negative nutrition densities, because a low price would make the food appear worse than a high price. To deal with this I shifted the nutrient density so that the worst food began at 0.
If you haven’t watched that presentation video above I encourage you to do so, it’s really interesting, and will explain the limitations of the nutrient density portion of this better than I can.
- The first main issue is that the nutrition data from the USDA isn’t complete, but it’s still the best source available. The main thing that it is missing is information about essential fatty acids, and some vitamins/minerals.
- The topic of enriched foods deserves a separate discussion in the near future, but obviously foods that are enriched will score very high on the nutrition side of this, which may or may not be fully deserved.
- Every region in the United States, let alone different countries may have drastically different prices for any number of foods. Buying in bulk vs. single, brand name vs. no name and during sales could affect results for your situation significantly. Feel free to download a copy at the end and put in custom data to see how foods you often buy stack up.
- While 1000+ entries is a decent sample, it is no means a complete sample and there are many foods I would have liked to have included.
There’s a lot that could be said about the information gathered, but I’ll try to go over the most interesting.
Here’s a quick overview of the distribution before we start looking at specifics:
There’s an average score of 25.68, and you can see that any score above 40 is pretty highly ranked.
The Kings and Queens of Nutrition and Cost
If you had a look at the beef cut nutrition profile tool that I posted a short while agoyou saw that the best part nutrition-wise was liver. It shouldn’t shock you that beef liver was in the top 10:
Most of the other top entries didn’t rank very high in terms of nutrient density, but were very cheap. I will come back to this to finish the article and try to answer the original question.
Spices have an interesting role when it comes to cooking nutritiously and cheap. With some practice, spices can make plain meals taste great, which may allow you to buy cheaper/plainer foods in the first place. So even if they don’t have particularly high scores, still may be worth buying.
They were of particular note in Lalonde’s presentation, where spices as a whole were a highly ranked category. I had a very limited sample of spices available, but they had a high cost/100 grams, which led to good, but not the best scores overall. The standout was Parsley, which had the 6th highest score.
Meat and Meat Products
Overall meat and meat products scored pretty well, which means they are often worth the extra expense over the cheapest of foods. Note that if this included essential fatty acids I would expect them to rank even higher.
Not surprisingly, the organ meats score very well as they are both nutritious and fairly inexpensive. Fish, ground meats, and sausages round out most of the other top spots for meat.
It’s interesting that the same types of foods that scored really well before score really poorly in a lower-quality form. This suggests that any savings from cheaper quality meat, particularly deli meats (like sausages) are not worth the nutritional trade-off.
Fruits and Vegetables
This is probably the section that surprised me the most, take a look:
In general fruits and vegetables both have poor nutrition densities. Sure we aren’t including fiber as an essential nutrient, but compared to the expectation of “fruits and vegetables are the healthiest foods ever”, it’s kind of a letdown. That being said, vegetables in particular are fairly cheap and you can see by the scores that they are pretty cost-effective.
Finally, while I was looking for Lalonde’s data I came across this reddit thread, and there was a comment I would like to relay here:
…For example, fruit and veggies scored very low in his list, but fruit and veggies have other types of benefits that go beyond vitamins and minerals (e.g. flavonoids, antioxidants, etc). Maybe their role is not to be nutrient dense, but to provide other functions for us.
I’d just like to re-emphasize that this isn’t the table to end-all nutrition arguments, but to provide a way to look at the cost-effectiveness of foods on an essential nutrient level.
In general grains are a very polarizing topic, which often brings up good discussion as a result. Anyone who has visited before probably knows I’m not a big fan of grains, which makes these results extra interesting:
Looking at those results we see that even though the nutritional content isn’t anything special, grains are just really cheap. On a tight budget, there is a solid argument to include them as part of a diet.
This brings up a very important limitation of this tool/table (or whatever you’d like to call it), which is that just because something has a high score doesn’t mean it’s appropriate for a high percentage of your diet. There are many issues that people commonly have with grains that may make it prohibitive to include any at all, even if they give you a decent amount of nutrition for your dollar.
Secondly, there is the issue of macro-nutrients, of which proper balancing is probably the most important part of a diet. What’s the point of getting nutrients if you’re killing yourself eating a disproportionate amount or carbs or protein?
I had to separate this from the rest of the grains for obvious reasons. The results are all pretty strong as well, owing to a decent nutrient density score and fairly inexpensive cost.
As I mentioned earlier, these cereals are obviously fortified with vitamins and minerals, which has a giant effect on nutrient density. The biggest question that surrounds these are the value/absorption of added nutrients.
Back to the Original Question!
Do you in fact have to resort to cheap and unhealthy foods while on a tight budget?
We saw that the highest scoring foods were predominantly due to being cheap, meaning you could eat a ton of them for the same amount as a more expensive food and overall get more nutrients. This is where caloric density must be considered, because eating 3000 calories in rolls would be ludicrous. Vegetables on the other hand may cost more per nutrient, but have a substantially smaller caloric density.
In my opinion, the spreadsheet/scores show that there are many foods that provide cost-effective nutrition. There are 132 foods with a rating over 40, and many of those also have a good nutrient density (the closest marker to a healthy food we have).
Keeping this in mind I believe that under the tightest of budgets it may be necessary to incorporate some less healthy foods that are cheaper, BUT in most cases it should be possible to construct a diet out of healthy foods with a high Nutrient Density/$ score.
If you’ve read all that thank you for your time and attention.
The original data sources were linked at the start of the article.
You can download a copy of my latest Nutrition Cost Score Table here to play with.
Note that the categorization was to help me find things and make simple comparisons, the sample size is much too small for any group vs. group comparisons.
I have some thoughts on how I’d like to develop this further, but I’d also like to challenge anyone reading to continue off on any tangents they find interesting/useful and let me know the results.
As usual, question or comment down below!