As Regras

It is always good to go back to first principles and look at what the rules actually say. From the Basic Rules (pp. 33-34) (Player's Handbook is identical I believe), I have highlighted what I consider to be key points:

Alignment

A typical creature in the worlds of Dungeons & Dragons has an alignment, which broadly describes its moral and personal attitudes. Alignment is a combination of two factors: one identifies morality (good, evil, or neutral), and the other describes attitudes toward society and order (lawful, chaotic, or neutral).

... Individuals might vary significantly from that typical behavior, and few people are perfectly and consistently faithful to the precepts of their alignment.

Alignment in the Multiverse

For many thinking creatures, alignment is a moral choice. Humans, dwarves, elves, and other humanoid races can choose whether to follow the paths of good or evil, law or chaos.

Alignment is an essential part of the nature of celestials and fiends. A devil does not choose to be lawful evil, and it doesn't tend toward lawful evil, but rather it is lawful evil in its essence. If it somehow ceased to be lawful evil, it would cease to be a devil.

The overarching concept is one of choice (unless you are a devil etc.); to my mind, the alignment written on a character sheet is aspirational - this is the type of person the character quer to be; not who they estão .

This contrasts sharply with the earliest editions of D&D where the consequences of consistently acting outside alignment were severe (loss of a level in AD&D - a fate worse than death for a character!)

The problem of Evil (and Good)

How do you recognize an evil (or good, lawful or chaotic) act? Are they objectively quantifiable or are they contextual?

The problem is akin to that addressed by Justice Stewart when considering pornography:

I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description [hard-core pornography]; and perhaps I could never succeed in intelligibly doing so. But I know it when I see it, ...

Similarly, evil (or good, lawful or chaotic) is hard to define but I know it when I see it!

Part of the problem is that even after several thousand years of thinking about this stuff, there is no consensus of what "good" is: is it altruism, utilitarianism, liberalism, egalitarianism, something else? Similarly with "law", do we need: democracy, communism, feudalism, socialism, something else?

I contend that the ethical content of an act is a combination of its consequences and the intenção of the perpetrator. This is reflected in most judicial systems: to be a crime the perpetrator must have criminal intent.

Under the RAW, an act done by an angel is both Lawful and Good simply because an angel is doing it. Equally, the mesmo act is both Chaotic and Evil if done by a demon for the same reason. It is within the DMs purview to decide that there may be acts that one can perform which the other is incapable of but I argue that these should be the exception rather than the rule. For example, both should be able to comfort or kill a child in the right circumstances; in either case an angel acting for the greater good and a demon acting on a titillating whim.

When the act is done by a PC, things really get confusing.

Reading the description of the alignments as they apply to those with moral agency, it is clear that you do not have to be vicious to be evil nor does being good mean you can't be vicious. As written, good is synonymous with altruism, evil with selfishness, law with community and chaos with individuality. An evil character may kill with regret, a good character with satisfaction, what matters is the motive for the killing.

As especificidades

The druid is acting bang on to his personality, he has no reverence for society, either the broader state or his current companions: chaotic to the core. Although he is clearly putting self interest first here, it is clear that he does not consider depriving his companions of their "fair share" to be either evil or good: in his mind it doesn't hurt them and he can use the money for useful stuff like saving the forest. The question of goodness comes when the fighter or paladin needs the money for something - will the druid give it to him?

You give no information about the the paladin except his alignment. He is clearly não acting to his alignment; he is not showing loyalty to the fighter (chaotic) and is acting in the paladin's own self interest (evil). Oh well, the paladin set high ideals for himself and failed to live up to them, don't we all? Maybe he will feel guilty about his small acts of Chaotic Evil, maybe he won't. Of course, maybe he doesn't like the fighter, maybe he feels he has earned the extra money, maybe he feels the fighter would just blow it on drink while he can use it to help the poor (that is, the poor who are not innkeepers).

Unless he is breaking his oath (which he isn't because he hasn't taken it yet) there are não in game consequences. You have some interesting opportunities, however, when he does reach 3rd level: perhaps he não pode take his oath. Some quests to find out why and to make amends might be in order, hmmm?

Resumo

To my mind, alignment for PCs is a role-playing guide like traits and flaws. I consider them as aspirations of who the character would like to be and I don't see it as the DMs role to castigate them when they turn out to be someone else so long as everyone is having fun.

If you really wan't to meddle in this then some positive reinforcement through inspiration when the character plays to their alignment at a cost é o caminho a seguir.

Mechanically speaking alignment has little to no bearing on play as per the 5e rules.

The PHB and basic rules mention alignment a lot, and its specifically described in detail in chapter 4 of the PHB, but how to implement it beyond a notation on a Player's character sheet is not given any space.

There a few optional rulesets in the DMG that hinge on using alignment to show corruption or character change over time, but again the mechanical impact is limited and not given anything but a brief description.

Ultimately alignment's impact is between you and your players.

As with many things int 5e, the impact of alignment and its importance to your campaign is largely dependent upon DM fiat and the social contract that is setup between the DM and the players. Alignment can serve as hard rules, limiting player actions lest they change alignment and lose access to certain features (as some previous editions handled it) or it can simply be used as a helpful guideline for good roleplay.

Addressing your specific situation

Ultimately morality is in the eye of the beholder, if you are looking to find out if the Druid's morality should change/his actions are against his alignment you should ask the player if the druid would think stealing/hiding a larger share of the loot would be considered wrong by his life experience and how he was raised. If his PC would see it as natural (working to help the greater good but also help himself in the process) then you should not instigate any alignment change. However he may get a reputation among groups and settlements for his greedy nature which may hamper his actions.

My Advice as a GM

Talk to your players out of character (all of them together) to see what their expectations are and where they would like the game to go. Ultimately whether you make Alignment have mechanical weight should be a decision made with everyone in mind, not one specific player or your own opinion. Personally I think 5e works best with alignment only serving as a moral guidestick to influence roleplay and that heavy mechanical weight hanging off alignment is best left to the previous editions that exercised it.

These are evil acts, but not necessarily out of character.

Stealing is chaotic, and stealing for your own gain (which both of these cases are) is evil. However, alignment does not dictate a character's actions, it reflects them. If your players frequently take these kinds of actions, you may well be justified in changing their alignment to reflect their true nature.

Exactly what changes you make to their alignment, and how drastic a change, depends on their actions. If the examples you've given are the extent of what they've done, and they've counter-balanced that with many more acts of selfless good, then their alignment might not change at all. If they do this kind of thing often, but also do acts of good in roughly equal measure, then their alignment might change to neutral. If these actions reflect their primary nature and there are few, if any, selfless acts to counter-balance it, they might well change to an evil alignment.

However, it is important to note that, just as a good alignment does not prevent a character from performing evil acts, nor does an evil alignment prevent a character from perform good acts. Your players' characters will remain the same characters, no matter their alignment. Changing their alignment to evil won't automatically cause them to start kicking puppies and killing people for no reason (evil is nothing if not reasoned). Neither will it cause people to run in fear, or try to kill them, or treat them any differently at all, unless their actions cause them to.

Alignment is, above all, meant to be an aid to role-playing, to help your players think more about their characters actions. Not acting within their alignment is not a failure, and does not deserve punishment. Their actions can, and often should, have consequences, but those should follow logically from the actions they're taking, not because they've failed to be the kind of saintly heroes you might have expected them to be.

Whether or not you obey the law in and of itself is measured on a scale of Lawful to Chaotic. Your motivations for (not?) obeying the law are what determines the Good to Evil axis.

An evil character can steal from the rich and give to the poor - not because he cares about the poor, but he really wanted to see how steamed his rival got when his precious gold was missing. A good character might steal someone's only horse because they need to either stop Baron von Evil with a stolen horse, or walk and be too late to stop the ritual. He might also steal the horse so that Plucky the Brash Level 1 Fighter doesn't try to fight a dragon and die senselessly.

The Good/Evil axis has muito little to do with law. Rather it has to do with prioritizing benefits for others vs. benefits for you. For example, a 'good' character will tend to prioritize the benefit of others over benefits for themselves. (Classic adventurer-tropes include risking your life to go deal with the goblin menace in order to save the village.) 'evil' characters will be much less likely to do so unless the benefit to themselves can be increased 'appropriately'.

Also, the Lawful/Chaos axis doesn't actually have much to do with following 'laws' per-se. Rather it has to do with following a code of conduct. (Laws are just an clear example thereof.) A 'lawful' paladin doesn't have to obey the tyrannical laws of the local lord if his code of conduct is to protect the weak. A 'chaotic' character will tend to act more on his/her whims.

Stealing (from the general public or from other PCs) isn't necessarily evil (if what and how much you steal does no harm, it's no worse than neutral on the good vs. evil spectrum), but it's certainly chaotic (putting your own needs or desires ahead of those of the group you've chosen to join). One can be genuinely evil and still lawful (a dedicated assistant to a genocidal dictator, like Himmler), one can be decidedly good and still chaotic (Robin Hood is a borderline example; Johnny Appleseed is probably a better one).

These actions could indicate a shift of the druid from chaotic good toward chaotic neutral, but I'd be more concerned about the (lawful good) paladin just going along with the actions -- that would seem like a shift toward both chaos and good vs. evil neutrality on his part (regardless of his state relative to vows he hasn't taken yet).

As many of the other answers pointed out there is no mechanical requirements to how you roleplay your alignment in 5E D&D. Previous editions included things like massive experience loss and other potential punishments as disincentives for alignment changes.

Stealing isn't really greed

Is the character pushing for missions/goals to make money? Is the character bargaining to get the better end of the deal? That's roleplaying greed.

If not, then it's not really about roleplaying as much as it is about the player selectively pushing the character to steal resources for game reasons. I've seen players do this out of some sense of fear, or lack of control (in how the game goes), or as some kind of competition against the other players.

In most cases where I have encountered this kind of behavior, that's what's going on. The easy solution is this: "Hey, you're doing this thing in play. Is this fun? Hey, other players, is this also fun for you? Does this fit with what we want to do?"

No, really let's play a game about moral quandries!

Now, if you want to really explore moral quandries about alignment, then everyone playing has to understand this is a key part of play and as a GM, you're going to be putting people into positions to test how they feel about their alignment stances.

"Hey Paladin, this kid is dying of a sickness that this medicine can cure, but they won't sell it to you, will you steal it?"

"Hey Druid, that necklace you stole turned out to be that woman's last keepsake of her dead mother. Are you gonna give it back?"

This can be a ton of fun, but D&D is not a game system particularly well set up for this and few players are able to deal with the mental shift from standard D&D expectations when you start throwing these problems their way.

I think a better question here would be, is "is being greedy and being good a good combination", as it has less to do with the lawful/chaos scale if you ask me.

Being greedy means you take things that other people possess because you want it, which is not always a problem on the law / chaos scale (a tyrant taxing poor peasants way too much to fill his own coffers is acting perfectly lawful, but he's still a jerk) but it is certainly a problem on the good/evil scale in most cases.

However, stealing is entirely a chaos/law scale slider. "I could use this to do more good than the paladin who will just donate it to his church anyway, so I'll keep it and lie" is lawfully dubious action, but not exactly evil, you're doing it because you intend to use the money for a good goal rather than because you like swimming around in it Scrooge McDuck style.

In the end though, remember that alignment, whether on the good/evil scale or the law/chaos scale is not something that is set in stone. There are plenty of good people with bad habits and plenty of villains with redeeming qualities and it doesn't automatically make them "neutral" simply because not ALL their traits are lawful/good/etc.

Greed is certainly at least a neutral trait; if it harms others then it starts to stray into evil territory. The essence of Good is generosity, just as the essence of D&D Evil is cruelty.

Stealing in itself can be good or evil - steal from the rich and feed the poor and you're good; steal from the poor because they can't afford weapons to stop you and you're probably evil.

The DM's responsibility is to track the character's actual alignment and adjudicate the effects. I would always warn a player if I felt that they were going to change alignment (in any direction), but the decision to change their behaviour would be theirs.

TL;DR: Deceiving party members and taking more than his fair share of loot is out of alignment for a good character, even if it is lawful for him to do so.

Goodness and Lawfulfulness are entirely seperate character traits, are therefore each is measured on its own axis: Good vs Evil, and Lawful vs Chaotic.

Being good does not necessarily incline a character to be lawful, and being lawful does not necessarily incline a character to be good. The same can be said for the opposite ends of the respective scales.

Some people have difficulty distinguishing between good and lawful for cultural socialisation reasons (in the real world we are taught to collapse this distinction), but nevertheless they are in fact unrelated.

Mal Legal

For example, it is entirely possible to be lawful evil. This type of character believes in order and authority above all else. He doesn't act out of "evilness", but out of self interest and to enforce the prevailing social order. Fictional example: Darth Vader (bringing order to the galaxy through ruthless oppression). The Lawful evil character is a common trope in film and television, and encompasses characters from bureaucrats to despots who work within the system to benefit themselves at the expense of others, either intentionally or through disregard.

In the real world, the exploitative (but law abiding) employer, the overzealous police officer, the soldier (essentially a mercenary for the state), the usurer (someone who lends you money knowing you will be trapped by the debt) or the member of an oppressive government (i.e. politician) are examples of lawful evil. They are highly immoral but obey the law and respect authority.

Caotic Good

At the opposite end of the spectrum is the chaotic good character. This is the Robin Hood archetype who does good independentemente of the law. He doesn't necessarily set out to break the law, but he's perfectly willing to do so to achieve a good outcome. Han Solo is another chaotic good character.

In the real world, the revolutionary fighting for freedom or justice, and citizens who engage in civil disobedience to protest some injustice are examples of chaotic good. You would find this kind of person working for Sea Shepherd or Greenpeace, or joining a revolutionary organisation. They are driven by morality and are willing to break the law if they feel it is unjust and doing so will lead to a better outcome for everyone.

O Druida

Aplicação de Perfeito is somewhat subjective to the character's values, however good aligned characters are generally concerned with the wellbeing of others and society in general, and hold honesty and loyalty to other good characters as important values. They are highly moral, and sometimes criminal.

Greed often leads to evil, but is not necessarily evil in and of itself. It's how the characters acts on his greedy impulses. However, taking more that their fair share and deceiving their party mates, even if it is perfectly legal to do so, is incongruent with a good belief system, and more akin to lawful evil than chaotic good.

You seem to be making the assumption that Lawful is the same as Good. Were it so, there could be no Chaotic Good, no Lawful Evil.

Imagine a society dedicated to the rule of law, upon which a charismatic political bloc comes to hold sway. This bloc utilizes some culturally compelling arguments to select a smaller subset of the population to serve as a scapegoat, to be sacrificed whenever things go sour. The populace and their leaders codify this into law, and suddenly a great injustice is lawful. That's Lawful Evil, and has occurred repeatedly throughout human history. Heck, if you look at history you can practically pick a year at random and this will have happened that year somewhere on earth.

The law is an ass, as George Chapman and Charles Dickens remind us, and Lawful Evil as well as Lawful Neutral are perfect examples of this.

Here's another example: Recall the American Revolutionary War wherein the colonists rejected the King's authority (ie laws) for what they certainly saw as the greater good. This could certainly be called Chaotic Good--an act completely violating the written laws, yet not to oppress any person or cause great bloodshed for its own sake, but to afford a great multitude of people a better life. Possibly it's Neutral Good, as they were not opposed to laws and good order. (If you disagree, again, hey, look, it's an illustration.)

There are a few methods suggested in the game books on how to handle this mechanically. Note I neither know nor care which ones--Rule 0 is RAW, the others are just inspiration.

One is to track deeds that violate the two alignment portions (Law/Chaos, and Good/Evil). After X steps away, the player gets a warning in some way (conscience, dreams, deities, etc). After a few more steps, their alignment changes--and any consequences which thereby result. For some characters that's nothing, and that's fine.

However, before even doing this, you should have a talk with your players on 1.) If you even want this in your game, 2.) How to determine these shifts, and 3.) What the impact is in the world. Most players don't seem to care much. Those players who choose a class with alignment restrictions generally care at least a little. Paladins and Clerics should care quite a bit... or at least their deities ought to!

Here's what I suggest for phasing in Alignment:

If your world is infused with magic (most are) but your players don't seem to care about alignment, you could go with a purely cosmetic result. Evil characters always manage to catch a shadow, there's a general feeling of ill will about them. Good characters may seem to glow and set people at ease. Lawful characters may resonate with people as trustworthy but a bit boring, while Chaotic ones excite people but don't exactly inspire confidence.

If they like this approach, start giving some mechanical modifiers. Bonuses to interactions, for example. If that goes well, grant an appropriate boon from a deity (their own or one whose alignment matches what the PC is moving to). A healing from the goddess of healing, perhaps, or a warning from the trickster god about a trap the PCs may run into. Let them know it's a deity intervening. Do it more and more as they come deeper into an alignment. Then...

...

... take it away when their alignment shifts. Leave a dead zone before another deity tries to curry their favour. That makes alignment something more than flavour text, but more than just a number and a bonus or penalty.

In general terms, lawfulness of an action by itself has no bearing on goodness. As it says in the PHB 122, alignment is a combination of two things. One is personal morality and the other is how they act in society and order. So, in terms of stealing, the reason behind the stealing is what matters to determine if the act is good or not, but not if it is lawful, which, unless you're talking about kinder society, it is not. So, in terms of your Paladin, it would be hard to justify stealing and lying. In fact, the first two oaths, Devotion and Ancients, either directly or indirectly prohibit stealing and the last oath, Vengence, could allow it only if it was directly necessary to exterminate his enemies. So, he could be an oathbreaker paladin technically and the DM should take a hand. As for the Druid, I suppose if he only cares about wildlife to the detriment of humanoids, then that would cover the smugglers if he puts the money to use feeding the animals or something. Honestly, it sounds like chaotic neutral would be a better fit for the Druid and I would probably suggest a change as the DM.