In Lima, Convenience Still Looks Like Food

What a city's markets, menus, and street kitchens reveal about the operating environment beneath our diets

August 16, 2026Read on Substack

When people in Málaga ask me where they can find good Peruvian food, I usually disappoint them.

Most of the time, I decline to recommend anywhere. Occasionally, I mention one or two places, but only for specific dishes. It is not because talented Peruvian cooks do not exist outside Peru. They do. It is because traditional food does not travel as easily as recipes may suggest.

A recipe can cross the Atlantic in seconds, but the ingredients cannot always follow.

The potato may be wrong. The corn may be too sweet. The peppers may look familiar but lack the necessary depth. The lime may be too mild, the fish too old, the herbs close enough in appearance but not in taste.

Substitutions can produce a good meal. Sometimes an excellent one. But not always the dish people believe they are eating.

There are foods I only cook when I return to Lima for that same reason. I know how to make them elsewhere, but the environment does not always provide what they require.

A recipe is information. A cuisine is an operating environment.

It depends on soil, climate, and biodiversity, but also on supply chains, markets, specialist retailers, cooking knowledge, and people who still know what an ingredient should look, smell, and feel like.

Returning to Lima makes that infrastructure visible because so much of it remains part of ordinary life.

The infrastructure between ingredients and meals

Across the Lima neighborhoods I know, neighborhood markets remain active parts of the food system.

People go there for vegetables, fruit, fish, poultry, and cuts of meat that are harder to find in conventional supermarkets elsewhere. The vendors do more than move inventory. They know the food. They can tell you what is fresh, what is in season, and what a particular cut is good for.

My local butcher sells me marrow bones for less than one euro a kilo. I try not to tell him what they charge for them as a delicacy elsewhere.

The difference is not contained in the bone. It is created by the system around it.

In one environment, marrow remains an ordinary by-product with a place in domestic cooking. In another, the supply chain and the knowledge around it have largely disappeared, allowing the same thing to return as a luxury.

The markets are only the first layer.

There are stalls inside and around them that turn those ingredients into breakfast or lunch. There are cevicherías, pollerías, and small neighborhood restaurants. There are carretillas, the carts and street vendors that have fed Lima in different forms for generations.

There are also huariques, modest, low-profile places often known for one dish or a particular cook. Some operate from conventional storefronts. Others began in markets, garages, or domestic spaces. Their attraction is rarely interior design. It is that someone knows how to make something worth finding.

And there is the menú.

A first course or soup, a main dish, and something to drink. Usually uncomplicated, typically affordable, and always delicious: cooked food intended to feed someone in the middle of an ordinary working day.

Spain has its own long tradition of menú del día. Málaga still has places where lunch means a first course, a second, dessert, and a drink rather than one assembled product eaten at a desk.

But I have only lived there for three years, and even within that limited period I have watched those places become harder to find.

Perhaps tourism is part of the pressure. Perhaps it is rising rent, labor costs, changing work rhythms, or the narrow economics of running a small restaurant in a city experiencing massive changes. Most likely, several forces reinforce one another.

I cannot establish a citywide trend from my personal field of view, but I can feel the layer becoming thinner.

In Lima, that layer still feels structural.

The menús, market counters, and street vendors are not experiences curated for visitors seeking authenticity. They remain ordinary interfaces between people and the food supply.

They allow someone to outsource cooking without necessarily outsourcing food itself to an industrial manufacturer.

That is worth noticing.

There is a difference between another person buying ingredients and cooking you lunch, and a corporation extracting, refining, reformulating, stabilizing, packaging, and marketing something designed to survive a global supply chain.

Both save time.

They do not produce the same relationship between the eater and the food.

More than four famous restaurants

Lima’s international culinary reputation is now easy to quantify.

In the latest published World’s 50 Best Restaurants list, four of the top fifty restaurants are in Lima: Maido at number one, Kjolle at nine, Mérito at twenty-six, and Mayta at thirty-nine.

But those restaurants are not the most important evidence of the city’s food culture.

They are its visible apex.

The more consequential achievement is that serious food remains available much further down the economic ladder.

The same city that sustains some of the world’s most celebrated restaurants can still feed someone from a handwritten lunch menu, a market counter, or a grill pushed into position on a neighborhood street corner.

These are not separate food systems.

The upper layer draws from the biodiversity, suppliers, techniques, and cultural memory maintained by the layers beneath it. Its chefs can reinterpret Peruvian ingredients because those ingredients remain alive in markets, households, and everyday expectations of taste.

A city’s culinary strength is not measured only by the heights its best chefs can reach. It is measured by how far down the price ladder recognizable food remains available.

Even the new Jorge Chávez airport reflects this cultural assumption.

The airport is polished and significantly more expensive than the city outside, but food is not treated as an incidental service tucked beside the gates. The departure experience is organized partly through the visual and social logic of a market: multiple counters, shared spaces, and different kinds of food presented side by side.

Lima carries its food culture all the way to the very end of your visit.

The more important question is whether it can preserve the ordinary infrastructure beneath the branding.

What the global map is actually showing

I was thinking about this when I encountered a report published in The American Journal of Clinical Nutrition.

The researchers assembled forty-seven country or regional dietary estimates, representing forty-six countries and one regional aggregate. They then matched forty-five of those estimates with World Health Organization disease data to model the burden associated with ultra-processed food consumption.

It is not a synchronized global census.

The underlying dietary surveys were conducted in different years. They did not all use identical methods, and some estimates describe specific populations rather than entire countries. The Ethiopian figure, for example, came from women of reproductive age.

Those limitations matter, but the map remains revealing.

The share of dietary energy coming from ultra-processed food ranged from 4 percent in the Ethiopian sample to 57.5 percent in the United States. The United Kingdom reached 56.8 percent. Across the high-income countries included, the average was 30.9 percent.

The study also modeled the health burden associated with consumption above a theoretical minimum of 10 percent of dietary energy, estimating approximately 3.4 million attributable deaths annually.

That figure should not be mistaken for a direct count. It is a comparative-risk estimate built from assumed dose-response relationships, a selected threshold, and existing cause-of-death data.

For the Human OS argument, the descriptive pattern is more important than the headline mortality number.

In some of the wealthiest and most technologically capable food systems in the world, industrial formulations now provide more than half of the energy people consume.

At that scale, “people should make better choices” is not an adequate explanation.

Americans and Britons did not independently decide, through millions of isolated failures of nutritional judgment, that most of their energy should come from ultra-processed products.

They adapted to an environment.

A food industry optimized for scale, shelf life, transportability, margin, and repeat consumption met a culture organized around speed, compressed schedules, portability, and minimal preparation.

The products fit the lives the system had produced.

What looks like personal preference at population scale is often infrastructure made visible.

The direction was visible years ago

The 2026 report did not reveal a sudden change. It consolidated a direction that has been visible for years.

A European study published in 2017 examined household food purchases across nineteen countries using surveys conducted between 1991 and 2008.

The data is too old to describe Europe today. They measured food purchased for the household rather than everything actually consumed, and they did not fully capture meals eaten outside the home.

But the variation was striking.

Ultra-processed products accounted for 10.2 percent of purchased dietary energy in Portugal and 13.4 percent in Italy, compared with 46.2 percent in Germany and 50.7 percent in the United Kingdom.

Countries exposed to similar technological development and regional economic integration had arrived at very different food environments.

That does not prove that markets, lunch culture, or culinary tradition caused the difference. But it does show that modernization does not produce one inevitable diet.

Some food systems retained stronger buffers than others.

The warning for Lima appears more clearly in a 2015 report from the Pan American Health Organization.

It examined sales of ultra-processed food and drink products across thirteen Latin American countries between 2000 and 2013. Per-capita sales increased by 26.7 percent across the region.

In Peru, they increased by 107 percent.

Sales are not the same as consumption. The categories covered by commercial datasets do not represent the entire diet.

But they reveal market penetration, and the report identified several forces accelerating it: urbanization, rising incomes, supermarket and convenience-store expansion, foreign investment, market deregulation, concentration among multinational manufacturers, aggressive marketing, and changing work and domestic rhythms.

The report found that sales volumes remained higher in wealthy countries, while growth was faster in lower-income countries whose markets had not yet become saturated.

That changes how I interpret what I see in Lima.

Perhaps I am not observing a permanent cultural difference between Peru and the countries where I have lived. Perhaps I am observing the overlap between two food systems.

One still depends on markets, cooks, street vendors, neighborhood restaurants, and ingredients that retain a recognizable relationship to the meal.

The other expands through standardized retail, packaged formulations, delivery platforms, and global supply chains.

There is one particular chain of convenience stores that appears to be increasing its presence in Lima fast enough to be visually noticeable between my visits.

They can coexist for a long time. Then one begins to displace the other.

The danger is not simply that people start eating more packaged snacks. It is that the people and places capable of supplying the alternative become economically unviable.

A market closes. A butcher retires. A lunch restaurant becomes a tourist concept, a franchise, or another short-term rental. A household loses the time or knowledge required to cook. A child grows up recognizing brands more easily than ingredients.

Each change appears minor.

Together, they alter the path of least resistance.

What the statistics struggle to see

Part of Lima’s food infrastructure is also difficult to capture because it operates informally.

Peru’s national statistics agency reported that informal employment remains widespread. In one INEI labor report from 2024, informality was above 70 percent in commerce and close to 59 percent across the broad services category that includes hotels and restaurants.

Those numbers do not tell us how many people sell lunches from their homes, operate street carts, or cook inside local markets. They do help explain why so much of what is obvious while walking through Lima remains faint in formal comparisons of food systems.

What is difficult to measure should not be treated as though it does not exist.

At the same time, informality should not be romanticized.

It can mean unstable incomes, weak labor protections, inconsistent sanitation, and limited regulatory accountability. The ability to enter the food economy with little capital often comes with personal and public costs.

But these small operators also constitute a distributed culinary capacity.

They are people who can turn ingredients into meals at a price and speed that make cooked food viable during an ordinary day.

When that middle layer disappears, the choice is no longer simply between cooking and eating out. It becomes a choice between doing all the work yourself and buying what industrial systems can most efficiently provide.

Ultra-processed food does not merely become tempting.

It becomes infrastructure.

The granny rule

After I shared an earlier essay about convenience with one of my communities in Málaga, one member offered a simple test:

Do not eat anything your grandmother would not recognize as food.

The objections arrived quickly.

“Ultra-processed” is a broad and contested category. Processing itself is not inherently harmful. Nutritional quality still depends on what is actually in a product.

All of that is true.

Humans have cooked, milled, fermented, brined, cured, and preserved food for millennia. Processing can make food safer, more digestible, more nutritious, and available beyond a short harvest season.

The concern is not that human hands or tools have altered an ingredient.

It is the growing distance between the ingredient, the final product, and anything that could reasonably be reproduced in a kitchen.

The granny rule is not good nutritional science. It is good interface design.

It gives an ordinary person a low-cognition way to navigate a commercial environment that has become increasingly difficult to interpret.

Most people cannot assess forty ingredients while standing in a supermarket aisle. They cannot determine what each modified starch, emulsifier, isolate, stabilizer, or flavoring contributes, much less evaluate their interaction.

They cannot audit the research behind every claim on the package.

A usable rule compresses that burden into a few questions:

Does this still resemble food?

Could someone outside an industrial supply chain make it?

Has processing served the ingredient, or has the ingredient become raw material for a formulation?

The rule will produce false positives. A grandmother from one culture may not recognize a legitimate food from another. Some modern formulations may be useful and defensible. Some traditional foods are nutritionally poor.

It is a rule of thumb, not a verdict.

But it also reveals something larger.

It is easier to follow in an environment where recognizable food remains accessible. It works better when local markets still exist, specialist sellers remain viable, and someone nearby can cook an affordable lunch from ingredients.

Even the best interface cannot compensate for an environment that has removed most of the acceptable options.

From nutrition to system design

The World Health Organization’s recent activity suggests that the institutional frame is beginning to widen.

In 2025, WHO began developing its first global guideline specifically addressing ultra-processed food consumption. In April 2026, it framed the subject through a One Health agenda connecting industrial food to agriculture, food safety, environmental sustainability, global trade, additives, packaging, and supply-chain governance.

That does not settle the debates around the NOVA classification, causal mechanisms, or where the boundaries of ultra-processing should be drawn.

It does recognize that the relevant unit of analysis is larger than the individual consumer.

Ultra-processed food is not merely a category of biological inputs. It is the output of a production system, a commercial system, and a way of organizing daily life.

There are at least three layers involved.

The first is food as input: what enters the organism and how it affects satiety, metabolism, energy regulation, and health.

The second is eating as behavior: what someone selects under the influence of taste, habit, price, stress, knowledge, and available attention.

The third is diet as system output: what becomes affordable, visible, portable, socially normal, and convenient within a particular environment.

Most nutritional advice concentrates on the first two. It analyzes the input and instructs the individual.

The third layer determines how difficult those instructions will be to follow consistently.

Lima reminds me that convenience does not have to mean industrial formulation.

A meal can be accessible without becoming unrecognizable. Cooking can be outsourced without surrendering nourishment to a factory. A city can make eating easier while preserving the relationship between ingredients, preparation, and culture.

But that capacity is not guaranteed in the long-term.

What I am seeing may be a temporary overlap rather than a permanent exception.

Lima can continue producing globally celebrated restaurants while the ordinary infrastructure beneath them gradually weakens. The top of the system may remain spectacular even as the middle becomes thinner.

The real question is not whether Lima can keep producing world-class chefs. It is whether markets, menús, butchers, cooks, street vendors, and neighborhood restaurants can remain economically viable parts of everyday life.

A recipe can be exported.

A food culture must be maintained.

A country does not arrive at a diet by making one choice. It builds a food system, a work culture, a commercial environment, and a set of daily constraints.

People adapt to those conditions. Eventually, the adaptation looks like preference.

This is why nourishment cannot be repaired through information alone. Food is an input, but diet is a system output. And when that output becomes pathological at population scale, the question is no longer only what people should eat.

It is what kind of life their environment has made possible.

If this helped you name something you have been feeling but had not quite put into words, please share it with someone who might need the same language.

The Human OS Manual is now available for pre-order. If this work resonates, you can reserve your copy and follow the ideas as they continue to develop.

Feel free to leave a comment. I read and respond to every thoughtful note.

Sources and notes